" :4 13"“.

u<unu
, .:.::
a..
3:!1‘5’ t
..x .. ....
4. .H. .:.I..............
a»:

.. :e.
1...:

“an. ”huh". 1.9.
. ti: 1...
l.

J”.

71...!
:52 .25.?
fix: 5.42:... .
E. i... i .
l _ .k a
a z :1.
.33....
. 3...:

.1. Ia.
1.. .9 .11

3...
2‘). .91:
15533.1...
x. .I .13.! l ‘Iug
o...
I. u z
..

i3} 173
21......

n. .
.......(|.. U
.3 V337...
a)....:........ :1
l u... .11! c.
3.1:... ...
c .1 t...
I 4...»: .
.31.... 011.10....
.I..v‘vh|:t
v.1.

.1
3 .1 I
:12... .

 

 

 

 

 

 

:3
.f7.

.3...

:Iv

93. . .32....

.vv. .1...
in... 1.!
{It}
4 2...?

.a
1...... 3...;

:5... 1

\ ‘1.‘ I
.. 1.5....
. up“. .

“Numb... 0

I...

‘i’i’
v 901'!
. x . ,
.v.»

.:

tub.-

 

”SSS

\JJ

It; ,1

lllllll'llllllllllllllllllllllll.lllllllllI

3 1293 01410 7258

This is to certify that the

dissertation entitled

INTERSPECIES GENE TRANSFER IN THE EVOLUTION OF
2 , 4-DICHLOROPHENOXYACETATE
DEGRADING BACTERIA

presented by

Catherine McGowan

has been accepted towards fulfillment
of the requirements for

 

 

Ph.D. degree in Microbiology

4222227612 é

Majo J‘protessor / f

   

Date é/{VI/Q 5/—

MSU is an Affirmative Action/Equal Opportunity Institution 0- 12771

 

 

 

LIBRARY 2
Michigan State :
University J

 

PLACE II RETURN BOX to roman this chockom from your ncord.
TO AVOID F INES mum on or More data duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU loAn Nflnnutlvo Adlai/EM Opportunity Intuition
Wan-9.1

 

 

INTERSPECIES GENE TRANSFER IN THE EVOLUTION OF
ZA—DICHLOROPHENOXY ACET ATE
DEGRADING BACTERIA

By

Catherine McGowan

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Microbiology

1995

ABSTRACT

INTERSPECIES GENE TRANSFER IN THE EVOLUTION OF
2,4-DICHLOROPHENOXY ACETATE
DEGRADING BACTERIA

By

Catherine McGowan

A great diversity of 2,4-D degrading bacteria exists in nature, yet it is
not known by what mechanisms this diversity evolved. This study has
demonstrated that interspecies gene transfer plays an important role in the
evolution of this diversity.

2,4-D degrading strains isolated from'various parts of the world were
analyzed for their phylogenetic diversity as inferred from their 168 ribosomal
DNA sequences. By this analysis, of 23 strains tested, 18 unique 16S rDNA
sequences were found which belonged to members of the alpha, beta and
gamma Proteobacteria.

The genetic diversity of the genes in these strains was further
examined by hybridization studies with tfdA, B, and C genes, which encode
the first three enzymes in the degradation of 2,4-D. Hybridization patterns
indicated fifteen different classes of 2,4—D degraders. These patterns also
showed that some strains contained pathways which appear to be mosaics of
tfd genes, which suggests interspecies transfer and recombination of tfd
pathways. This hybridization analysis also showed that two or more tfd
probes hybridized to all beta and gamma Proteobacteria, but not to members
of the alpha Proteobacteria. Only tde could hybridize at low stringency to

only some members of the alpha Proteobacteria.

To compare the phylogeny of tfd sequences with the phylogeny of 16S
rDNA sequences, tfdA genes from the beta and gamma Proteobacteria were
partially sequenced. Three classes of tfdA-like sequences were found in these
strains, differing by up to 24 percent from the corresponding sequence for
IMPl34. Dendrograms derived from these tfdA genes differed dramatically
from the dendrogram of the 16S rDNA sequences from the same organisms,
providing compelling evidence of interspecies horizontal gene transfer.

To discover whether these strains could mobilize 2,4-D genes in vitro,
they were mated with recipient strain Burkholderia cepacia D5. Several
strains transferred and expressed the ability to degrade 2,4-D, whereas other
strains did not. Those that rapidly transferred and expressed the 2,4-D
phenotype included strains with tfdA genes greater than 99% similar to tfdA
from IMP134. Those strains that did not transfer the ability to degrade 2,4-D,
or did so only at extremely low frequencies, had tfdA gene sequences differing
by almost 24 percent from IMP134. One of these non-conjugal 2,4-D strains,
Burkholderia strain RASC, was found to transfer its 2,4-D genes from its
chromosome into the chromosome of B. cepacia at an extremely low

frequency and required a long incubation period.

ACKNOWLEDGEMENTS

This dissertation was completed with the help of many people, whose
efforts I would like to acknowledge here. I thank my advisor, Dr. Jim Tiedje,
for giving me freedom to pursue my ideas and for his teaching that in science,
as often in life, there are no problems, only challenges. I also thank my
committee members, Dr. Michael Bagdasarian, Dr. Wendy Champness, and
Dr. Pat Oriel for their help and encouragement. I also thank Dr. Richard
Lenski for agreeing to be on my committee at a late date and for encouraging
me to compare phylogenies of tfdA with that of the 16S rDNA sequences. I
also thank Dr. Lynn Margulis for fanningthe flames of my curiosity about the
evolution and diversity of life.

For providing the strains used in this study, I would like to thank Dr.
Nancy Tonso Sasaki, Dr. Bill Holben, Dr. Jong-Ok Ka, and Dr. Roberta
Fulthorpe. I would especially like to thank Dr. Fulthorpe for her help in
growing about half the strains, extracting DNA from them, and checking for
2,4-D degradation. I thank Arturo Massol-Deya, for the rDl and fD1 primers
used in PCR amplification of the 16S rDNA genes. I also thank the Michigan
State University DNA synthesis laboratory for synthesizing the 519R primer
used in sequencing and Sue Lootens from the MSU molecular sequencing
facility for sequencing the genes on the ABI automated sequencer. For strain I-
18, I owe thanks to Dr. Olga Maltseva for preparation of this strain and its
DNA and assistance in PCR amplification and sequencing of this strain. I owe

iv

many thanks to Dr. Tom Schmidt and Dr. Jim Smith for the useful
discussions about data analysis. I thank Randy Russell for assistance in
programming a distance matrix algorithm for analyzing sequence data. I am
grateful to Dr. Cindy Nakatsu for teaching me how to use GenBank and
BLAST. I also thank Dr. Jim Cole, Amy Foxworthy, and Dr. Jizhong Zhou
who shared their ideas on acquiring the sequence data.

I give special thanks to Dr. Roberta Fulthorpe who collaborated with me
on the hybridization work and who will be first author on a publication
resulting from this work. I thank Dr. Bill Holben who gave us DNA for
making probes from tfdA, B, and C for the hybridization studies.

For the design and preparation of the tfdA primers, I owe thanks to Dr.
Tatiana Vallaeys and Dr. Alice Wright. Dr. Fulthorpe, Dr. Wright and I
together used these primers to PCR amplify the sequences. Dr. Wright will be
first author on a publication about the tfdA sequencing. I owe thanks to Dr.
Olga Maltseva for assistance in the preparation of tfdA sequence from strain 1-
18. For tfdA sequence information, I thank Dr. Yuichi Suwa for RASC, Dr.
Grace Matheson for TFD6, Dr. Cindy Nakatsu for T'FD41 and Dr. Fumiyasu
Fukumori for TFD159.

I owe thanks to Dr. Yuichi Suwa for the preparation of a probe from
RASC. I also thank Dr. Larry Forney for encouraging me to pursue the gene
transfer experiments from strain RASC.

I especially want to thank Fly for his loving support and discussions
about science. I thank Supai for her ever wagging tail.

TABLE OF CONTENTS

List of tables ......................................................................................................... vii
List of figures ....................................................................................................... viii

Chapter 1. Literature Review: Gene Transfer in the
Evolution of 2,4-Dichlorophenoxyacetic
Degrading Bacteria ......................................................................... 1

Chapter 2. Phylogeny of 16S rDN A Genes from Diverse
2,4-Dichlorophenoxyacetate Degrading Bacteria .................... 50

Chapter 3. Diversity of 2,4-Dichlorophenoxyacetate Degrading
Bacteria Determined by Gene Probe
Hybridization Analysis ................................................................ 94

Chapter 4. Interspecies Gene Transfer in the Evolution of
2,4-Dichlorophenoxyacetic Acid Degrading Bacteria
in Nature and in the Laboratory ............................................. 115

Chapter 5. Interspecies Transfer of Chromosomally
Encoded 2,4-Dichlorophenoxyacetic
Acid Degrading Genes ............................................................... 145

Bibliography ......................................................................................................... 176

vi

LIST OF TABLES

Table i Page
Chapter One

Genes from JMP134 involved in 2,4-D degradation .............................................. 17

Selected list of 2,4-D degrading organisms ............................................................... 19

Nomenclature key for class Proteobacteria ............................................................. 32

Similarity of strains comparing 168 sequence identity

versus percent of DNAzDNA hybridization ........................................................... 29

Recipients of plasmid pJP4 from Alcaligenes eutrophus JMP134 ...................... 42

Gene exchange in nature ............................................................................................. 45
Chapter Two

Strains used for sequencing of 16S rDNA ................................................................ 60

Presence or absence of restriction fragments of amplified 16S rDNA ............... 67

Reference strains used in building phylogenies ..................................................... 79

Distance matrix indicates percent sequence similarity based

on 283 aligned nucleotides .......................................................................................... 78
Chapter Three

2,4-D degrading strains, their origins, references and REP groups ................... 101

Hybridization patterns of 2,4-D degraders .............................................................. 103

vii

Chapter Four
Strains used for comparison of tfdA and 16S rDNA sequences ........................ 123

Distance matrix of tfdA-like genes from strains JMP134,
TFD39, TFD23, RASC and TFD31 ............................................................................. 127

Strains used for mating studies and results of in vitro
2,4-D phenotype transfer ............................................................................................ 134

Chapter Five

Conditions used in matings ...................................................................................... 161

viii

LIST OF FIGURES

Figure Page
Chapter One
Three models diagraming possible evolution of 2,4-D degraders ......... .4
Phylogenetic congruency test .......................................................................... 6
Systematic phylogeny of all life from origin to present .......................... 12
Pathway of 2,4-D degradation in Alcaligenes eutrophus JMP134 ......... 16
16S rRNA universal phylogenetic tree ...................................................... 24
Diagram of 16S rRNA molecule .................................................................. 25
Comparison of genes for ortho cleavage .................................................... 35
Diagram of interspecies gene transfer ......................................................... 37
Plasmid pJP4 from Alcaligenes eutrophus JMP134 .................................. 40
Chapter Two
ARDRA analysis of 2,4-D degraders ............................................................ 61

Dendrogram of 2,4-D degraders based on Maximum likelihood
analysis of restriction data ............................................................................. 68

Flowchart indicating the collection
and sequencing of 165 sequence data .......................................................... 70

Chromatogram of Applied Biosystems model 373A automated
sequencer ........................................................................................................... 71

ix

Chapter Two (continued)

Alignment of 16S sequences from position 209-506
according to the E. coli numbering system ............................................... 72

Dendrogram of 2,4-D degrading strains and reference

strains based on Neighbor Joining Method of PHYLIP ........................... 80
Phylogeny, based on maximum parsimony, of 16S rDNA

from 2,4-D degraders ....................................................................................... 81
Geographic distribution of 16S rDNA ......................................................... 92

Chapter Three

Strategy for hybridization at high, medium and low stringencies ..... 102
Hybridization patterns with tfdA, B, and C probes

arranged in phylogenetic grouping ........................................................... 105
Hybridization patterns of tfdA, B, and C probes from pJP4 arranged in
phylogenetic groupings ................................................................................ 107
Mosaic pattern of tfdA genes indicates recombination ......................... 111

Chapter Four

Alignment of 313 nucleotides of internal fragments of

tfdA genes from representative strains .................................................... 125
Dendrogram of 2,4-D degrading strains based on tfdA sequences ...... 128
Geographic distribution of tfdA genes ...................................................... 129

Phylogenetic incongruency for tfdA genes and 16S rDNA
from diverse 2,4-D degraders ...................................................................... 130

Comparison of tfdA genes to phylogeny of 16S rDNA
from 2,4-D degraders and reference strains .............................................. 132

X

Chapter Four (continued)

Colony blot of recipients, donors, and transconjugants
hybridized with plasmid pJP4 ..................................................................... 136

Chapter 5

Southern blot of EcoRI-digested total genomic DNA from
RASC, B. cepacia D5, B. cepacia D5(RASC) and
JMP134 hybridized with tde ...................................................................... 156

DNA from JMP134, RASC, B. cepacia D5, and
B. cepacia D5(RASC) hybridized with tfdA from RASC ....................... 157

Duplicate colonies of RASC, B. cepacia D5 and B. cepacia D5(RASC)
hybridized to 11 kb plasmid from strain RASC ...................................... 158

Total genomic DNA from RASC and D5(RASC)
digested with seven restriction enzymes and
probed with tfdA from RASC ..................................................................... 160

Total genomic DNA from B .cepacia D5, RASC and three separate
transgenic strains digested with BgIII and hybridized with tfdA from
RASC ................................................................................................................ 163

Strains RASC, B. cepacia D5 and seven transgenic strains
PCR amplified using REP primers ............................................................. 165

REP-PCR Southern blot from RASC, B. cepacia D5 and
seven transconjugants and probed with 1.2 kb REP fragment from
RASC ................................................................................................................ 166

Hybridization of EcoRI digested total genomic DNA
with REP-amplified fragment from RASC .............................................. 167

xi

Chapter One

Introduction:

Gene Transfer in the Evolution of 2,4-D Degrading Bacteria

Chapter 1. Introduction: Gene transfer in the evolution of

2,4-D degrading bacteria
TABLE OF CONTENTS
OVERVIEW ....................................................................................................................... 3
RATIONALE ...................................................................................................................... 8
Bioremediation ...................................................................................................... 9
EVOLUTION
Gene transfer in evolution ...................................................................................... 10
Evolution of pathways for biodegradation of
xenobiotic compounds ......................................................................................... 11
2,4-D MODEL SYSTEM
2,4-D degradation ................................................................................................. 14
Known 2,4-D degradation pathway ....................................................................... 15
DIVERSITY OF 2,4-D DEGRADING ORGANISMS ............................................................ 18
Phenotypic classification ...................................................................................... 20
Phylogenetic classification ................................................................................... 22
16S rRNA ............................................................................................................. 23
Inference of Phylogeny
Parsimony ................................................................................................ 26
Distance matrix ........................................................................................ 27
Maximum likelihood ................................................................................ 27
Confidence test for phylogenies ................................................................. 27
Phylogenetic Congruency................. ...................................................................... 28
Species definition ................................................................................................. 28

DIVERSITY OF 2,4-D DEGRADING GENES

Gene similarities .................................................................................................. 33
Other chloroaromatic degradation genes ............................................................... 34
GENE TRANSFER POTENTIAL
Horizontal Gene Transfer ...................................................................................... 36
Conjugation ........................................................................................................... 38
Plasmid pJP4 ........................................................................................................ 39
Other 2,4-D plasmids ........................................................................................... 41
Chromosomally encoded 2,4-D genes ...................................................................... 43
Gene transfer in the environment ........................................................................... 44
SUMMARY ................................................................................................................ 47

OVERVIEW

Interspecies genetic transfer may play an important role in the
evolution of diversity of 2,4-D degrading strains. Gene transfer has been
shown to be involved in the evolution of xenobiotic degrading organisms in
vitro (Kellogg et al. 1981; Latorre et al. 1984; Reinecke and Knackmuss, 1979),
yet it is not known how important gene transfer is in nature for the
evolution of strains capable of degrading xenobiotic compounds.
Hypothetically, the diversity of organisms capable of 2,4-D degradation could
have evolved in one of three ways either by 1) divergence of an existing strain
into unique strains of 2,4-D degraders; 2) convergence or parallel evolution of
two unrelated strains into 2,4-D degraders; or 3) gene transfer, in which the
genes for 2,4-D degradation evolved once or more times and were
subsequently transferred between unrelated strains (Figure 1). Of course,
there may also be many permutations to these three basic themes. For
example, only the upper pathway may have transferred to strains which had
already evolved a lower catechol dioxygenase pathway. The goal of this
project was to examine the hypothesis that gene transfer played an important
role in the evolution of diversity of 2,4-D degraders.

Although no one argues that genetic transfer occurs in bacteria, and
that interspecies gene transfer has been observed in the laboratory, the view is
generally held that gene transfer in nature generally occurs within a species,

(Dykhuizen and Green, 1991) and that the range of gene transfer even defines

+
+
I
I

Divergence

Convergence

Gene Transfer

Figure 1. Three models diagraming possible evolution of 2,4-D
degrading strains. Stippled lines and a ”+" indicate the presence of
2,4-D pathway. Solid lines and a ”- indicate strains that do not have
the 2,4-D pathway.

5

a species, analogous to the definition of species in multicellular, eukaryotic
organisms (Stackebrandt et al. 1991).

One way to determine whether gene transfer is responsible for the
evolution of 2,4-D degrading strains is to apply a "phylogenetic congruency
test" that predicts if the phylogenies of two separate genes are different
(incongruent), then horizontal gene transfer has occurred within the group of
organisms under investigation (Wilson et al. 1977). In other words,
"recombination mixes the phylogenetic relationships of the strains"
(Dykhuizen and Green, 1991) (Figure 2).

In this study, I compared the phylogeny derived from 165 gene
sequences to the phylogeny derived from 2,4-D genes, in particular the tfdA
gene, the first gene in the pathway of 2,4-D degradation. Then, to see
whether transfer of the 2,4-D genes was observable in vitro, these strains were
subjected to mating experiments. The gene transfer of one strain was then
studied more in depth.

To accomplish these tasks, I performed the following experiments
which are described in the next four chapters. In the second chapter, I
examined the phylogeny of a group of 2,4-D degrading organisms based on
partial sequences of 16S ribosomal DNA. In the third chapter, I describe work
done by Dr. Roberta Fulthorpe and myself on the diversity of 2,4-D degrading
genes from diverse bacteria that hybridize to the tfdA, tde and tde genes
from plasmid pJP4. In the fourth chapter, I describe partial sequencing of the
tfdA genes from a select group of 2,4-D degraders and comparison of the
dendrogram derived from this information to that derived by sequencing of
the 16S rDNA genes. Finally, in the fifth chapter, I describe an unusual 2,4-D

Gene X

A B C D E
Gene Y

A D B C E
Gene Z

Figure 2. Phylogenetic congruency test. If phylogenies for two genes are identical,
or congruent, as are gene X and gene Y in this diagram, then gene transfer has not
necessarily occurred. However, if phylogenies for the two genes are different in
the same group of organisms, then this is evidence of horizontal gene exchange, as
is the case illustrated here where the phylogeny of gene Z differs from those of
gene X and gene 2.

7

gene transfer dynamic in strain RASC.

Before describing these experiments, I use the rest of the introductory
chapter to provide’background information on the project. I describe the
rationale for this project; give an overview of the evolution of xenobiotic
degradation pathways; describe the 2,4-D model system, the diversity of 2,4-D
degrading organisms, and their genes; and I describe mechanisms of gene

transfer and discuss what is known about gene transfer in the environment.

RATIONALE

"The great aim of education is not knowledge, but action"

~Herbert Spencer (1820-1903)

Earth has entered a new geological era, that of the psychozoic, in which
human influence on the biogeochemistry of the planet approaches the
magnitude of previous abiotic (tectonic) and biotic (i.e. production of
oxidizing atmosphere) forces (Piruzyan et al., 1985). Humans have influenced
the biogeochemistry of the planet in two fundamental ways: by the generation
of millions of axenic compounds and by an explosion in population growth
which in turn has accelerated and altered many biogeochemical processes.

One of these influences is that humans synthesize most of the two
hundred thousand novel substances added yearly to the list of over 5 million
chemical compounds described in Chemical Abstracts (Piruzyan et al., 1985)
and the United States alone produces approximately 100 million tons of
synthetic organic chemicals annually (Cork and Kreuger, 1991). What is the
fate of these xenobiotic compounds in nature? Some, usually analogs of
natural products, may be degraded by extant microorganisms, while for some
others, microbes may evolve novel pathways to degrade them (Allewell, 1989;
Yano, 1980, Omston and Parke, 1976).

One class of anthropogenic compounds is pesticides, used to combat

plant pests which destroy nearly one-third of the world's agricultural output

9

(Davidson et a1. 1991). In the United States approximately 200,000 tons of
herbicides are applied annually at a cost of about $5 billion (Zilberman et al.,
1991). The initial cost of pesticides, which have allowed a "Green
Revolution" enabling farmers to grow food enough to feed the planet, does
not include the potential hidden costs, such as contamination of the
environment (Postel, 1988). The removal of unwanted pesticides in some
cases may be achieved through biodegradation. Some pesticides are known to
be degraded due to the evolution of microorganisms capable of metabolizing
the compound as a substrate for growth. A factor contributing to this
metabolic evolution is thought to be the transfer of pesticide degrading genes
between various microbes (Fox, 1983; Kellogg et al., 1981; Yano, 1980).

To fully understand how microbes have evolved to degrade xenobiotic
compounds will require a substantial effort by many research groups.
Nonetheless, one potentially important component of microbial evolution, i.
e., the role of gene transfer, was examined using the pathway for the
degradation of 2,4-dichlorophenoxyacetic acid as a model system.
Bioremediation: The natural phenomenon of biodegradation is now being
exploited by humans to remove toxic, manmade chemicals from chemical
spills, aquifers, and Superfund sites (Kobayashi and Rittman, 1982). An
estimated 20% of the 1245 Superfund sites, for which the US. federal
government spends approximately $11 billion annually to clean up, might be
detoxified biologically (Begley and Waldrop, 1989; Ghisalba, 1983, Ghosal et a1.
1985, Gibson, 1984; Madsen, 1991; Reinecke and Knackmuss, 1988; Rochkind et
a1. 1986). Various strategies have been employed to facilitate biodegradation,

including addition of nutrients or aeration to encourage growth of

10

indigenous microbes, as done with the Exxon Valdez oil spill (Fox, 1989) or
addition of exogenous microorganisms capable of utilizing the substrate
(Roberts, 1987). In'the future, genetically engineered microorganisms may be
useful for some bioremediation. If the release of genetically-engineered,
detoxifying microorganisms (GEMS) is approved, introduction of degradative
genes on broad host range, conjugal plasmids or otherwise mobilizable genes
may be an effective method for bioremediation, since the pathways effecting
the metabolism of recalcitrant compounds may function best if naturally
transferred into indigenous populations at the site instead of by adding a
foreign strain which would be less likely to survive (Goldstein et a1. 1985).

With more knowledge of how catabolic strains have evolved in
nature, humans can use a more holistic approach to environmental

bioremediation (Verstraete and Top, 1992).

EVOLUTION

"The extent to which we can hope to understand
ourselves and to plan our future depends in some measure on our
ability to read the riddles of the past."

~ George Gaylord Simpson

When and how did 2,4-D degrading microbes evolve? If we can learn
how 2,4-D pathways and organisms evolved we might develop better
solutions for pollution management.

Learning about the mechanisms of evolution in a relatively simple

system such as 2,4-D degrading bacteria can also shed light on evolutionary

11

processes in general and add to our understanding of the evolutionary
relatedness of all organisms (Figure 3). A firmer understanding of
evolutionary processes may bolster the theory of evolution, which nearly fifty
percent of the U. S. population doubts, in favor of a theory of creationism
(Gallup, 1994, Taylor, 1994). An excellent model to understand the
fundamental principles of evolution is in the evolution of the diversity of
2,4-D degrading bacteria.
Evolution of pathways for biodegradation of xenobiotic compounds: Gene
transfer has been shown to be involved in the evolution of novel pathways
by the recruitment of protein modules (Baron et al. 1991). Organisms with
novel catabolic pathways appear to have evolved as the result of gene transfer
in vitro (Kellogg et al. 1981; Latorre et al. 1984), yet it is not known how
important gene transfer is in nature for the evolution of novel organisms.
It is surprising that so many exotic, anthropogenic organic compounds
can be biodegraded since microbes were presumably not exposed to these
compounds in any great quantity prior to the 1940's when they were first
synthesized and therefore might be expected to lack enzymes capable of
recognizing such compounds as substrates. Even though over 1500 naturally
occurring chlorinated compounds (including 2,6-dichlorophenol, the sex
pheromone of the female lone star tick ) have been isolated (Suide and
DeBernardis, 1973; Rhodes, 1994), no naturally occurring chlorinated
phenoxyacetic acid has been reported. Have organisms evolved catabolic
pathways de novo since the widespread introduction of these artificially
synthesized compounds, or is this simply outgrowth of already existing

pathways?

12

EONS

sawouafi a|6ugs
euaioeqaeqmv

9an

at”

sawouefi

sawouefi ajfiugs

"3 \

5810

8918:38an r—_

.83 £282 89c wagons 458.2“ 2 Emma 69w mm: an we fame—Fm omen—~33 .n mama

 

Roget
ammo
3663 U

DENEEEQLEQ .. EEESEQ. 4

 

$88331

 

6.68nger 263.3% _ combat 6285 m

 

 

 

 

 

 

 

.qqu.

o.

        

    

     

so: no

g.“

C .
coo-coo...
00......-
o o o .
C C .
0.0
o... o o
noon...
on. o
.- goo.
cue-u...
no.

   

 

dqfiflfidfidqqqu—u qu-uqqquu.ququ--1d-q

3553.8 ”may”.

$359058: : a a a

/ X\ Emufiofimoimfimg

\\ .
t... $883353 0
n 3353858 w

      

9 ON mm on on oe me .312.» .0. J
$553.95

 

I

I

<mm202

 

on
gm» 1 99 /0
é

o
.1
99.0

m . .........r

 

 

13

The research described in this dissertation has explored the possibility
that novel 2,4-D degrading organisms have evolved via gene rearrangement
and recruitment facilitated by gene transfer. Several researchers have
explored the evolution of metabolic function, which is reviewed by Mortlock
(1992); Omston and Parke (1976); and Schléimann et al. (1992). Evidence for
evolution by gene exchange comes from laboratory experiments (Kellogg et
al., 1981; Latorre et al., 1984) but is thought to occur in nature as well.
Although scientists have described the prevalence and importance of gene
transfer in the environment (for review see Coughter and Stewart, 1989; Fry
and Day, 1989; Levy and Miller, 1989; Levy and Marshall, 1988; Stotzky, 1989;
Stotzky and Babich, 1986; Trevors et a1. 1986) it is still not known to what
degree gene transfer in the environment facilitates the emergence of novel,

xenobiotic-degrading strains.

14

2,4-D MODEL SYSTEM

2,4-D degradation: A useful model for studying evolution of catabolic
pathways is the pathway for 2,4-dichlorophenoxyacetate degradation. One
reason for its usefulness as a model is because 2,4-D was presumably not a
substrate for bacteria since first being synthesized by humans in 1942 (it has
never been detected in pristine environments). Other reasons for choosing
2,4-D biodegradation as a model system include its widespread use, that it is a
synthetic compound of moderate recalcitrance, and because one pathway for
its degradation has been characterized biochemically (Duxbury et al., 1970;
Tiedje et al., 1969; Bollag et al., 1968; Loos et al., 1966; Don and Pemberton,
1991) and genetically (Perkins et al., 1990; Kaphammer et al.,1990; Don and
Pemberton, 1985; Fukumori and Hausinger, 1993; Suwa et al. in prep;
Matheson et a1. 1994; Nakatsu et al. in prep; Kamagata, et al. in prep).

The herbicide 2,4-D is a functional analog of the plant hormone, auxin,
that kills dicotyledonous plants by causing them to ”grow to death” (Corbett,
1984). Over 60 million pounds of 2,4-D (sold as over 1500 different registered
products including the bestselling lawn herbicide "Weed-B-Gone," (Ortho))
are applied annually for agriculture, in forests and on turf to control broadleaf
weeds. This herbicide also was used as a defoliant during the military
engagement in "Operation Ranchland" in Viet Nam. The Environmental
Protection Agency (EPA) has concluded that 2,4-D poses no significant health
hazards when used according to the label. This limited hazard is due, at least
in part, to the fact that 2,4-D is readily degraded in the environment.

15

Known 2,4-D degradation pathway: The pathway of 2,4-D degradation
encoded by plasmid pJP4 in Alcaligenes eutrophus JMP134 is the most well
characterized of all 2,4-D pathways (Figure 4) and its enzymes are best
understood (Table 1). The initial reaction is catalyzed by the product of the
tfdA gene, 2,4-D dioxygenase, an alpha keto glutarate dependent dioxygenase
(Fukumori, and Hausinger, 1993) which cleaves the ether linkage yielding
glyoxylic acid and and 2,4-dichlorophenol. This latter compound is
subsequently hydroxylated by 2,4-dichlorophenol hydroxylase, encoded by the
tde gene, to form 3,5-dichlorocatechol (Perkins et a1. 1990). This ring
compound is then cleaved via a modified ortho pathway, to cis,cis-2,4-
dichloromuconic acid, and then to 2-chloro-4-carboxymethylene but-2-enolide
and finally to 2-chloromaleylacetic acid (Perkins et al. 1990). Chromosomally
encoded enzymes further metabolize 2-chloroma1eylacetic acid to succinate
and acetyl CoA (Kukor, et al, 1989).

Although only one pathway for the degradation of 2,4-D has been well
characterized, other pathways may exist (as have been shown for toluene
degradation) but have not yet been described. An additional pathway for 2,4-
D degradation has been partially characterized which includes the formation
of a 3,5-dichlorohydroxyhydroquinone intermediate in Nocardioides simplex

(Kozyreva et a1. 1993).

:HZC' ZCOOH

16

,:°°°\
\
Cl

 

 

:' tde
2.4-0 la-ketoglutarate j trans-chlorodiene-lactone
dioxygenase isomerase
i c: \ofl/COOH
('3: W3 tde
j 2.4-dichlorophenol \j chlorodienelactone
hydroxylase hydrolase
HOOC
“6’ ”11
cu \ \0
mar
chlorocatcechol- j chloromaleylaoetate
1 ,2-dioxygenase reductase
HOOC HOOC
\/C| \
\ .. “U
c, chloromuoonate
cycloisomerase

Figure 4. Pathway for 2,4-D degradation found in Alcaligenes eutrophus
JMP134. (Diagram courtesy of Dr. Cindy Nakatsu).

17

mafiuuarmxocosm u <43 ..

$2 .a a 843$ 69.: £85 98 agnmfinv awe :3 8 99.3. 6 “$2 :3 z... 559% S «82 .1 s .9835 an "88m

 

 

u an: Co 8336: .8859: .533»?— my:

a 8.2:? 25.332 Ea. Ea 82. .5... .o 8:3»: «a.

u 638:... 338085920 38928.. 2232.338 .3...

a 92 82 822:8. 88.35.33.928. as:

a 0.: 8296.3 «cote—05G max.

a 8.95028“. baaozstom 95‘ 2: 88060306.? 23855835 Dub

a EU as savages 3 _2§32o2.... ol.

9 =8 .m 5 832%. 82 95. 8: 3...»..82; .ocofieo_fi_3.~ £5

a a 25 8 538» $28 <8 "8 Savage... u 8.55» 0.5 55
.8853— .o.oz .830 «in ma fig «50 88

 

£3833 9.5 5 8235 SEE 58. 880 A ~33.

18

DIVERSITY OF 2,4-D DEGRADING ORGANISMS

Before examining the role of gene transfer in generating diversity of
2,4-D degraders, the range of different organisms capable of degrading 2,4-D
needs to be presented. Understanding the diversity of bacteria is not a simple
task, since there are different and sometimes conflicting measures of
diversity, different schemes used to classify this diversity, and even different
definitions for the word "species" used to describe this diversity. In this
section I will discuss the diversity of 2,4-D degrading organisms already
known. Then I will briefly discuss the methods used to characterize and
classify the diversity of recently isolated 2,4-D degraders in an attempt to give
them a species designation. In particular, I will emphasize the use of 16S
rRNA as a tool in determining phylogenetic diversity and mention some of
the methods used to infer phylogenies based on 16S rRNA sequences. Then I
will discuss potential problems in defining diversity, including ambiguities in
nomenclature and the definition of species.

Many organisms have been isolated that can grow on 2,4-D as sole
carbon and energy source, including Arthrobacter (Loos et al., 1966; Tiedje et
a1. 1969); Bordatella (Greer et a1. 1992); Plavobacterium, (Chaudry and Huang,
1988); Alcaligenes, Pseudomonas cepacia (Greer et a1. 1990), and others (Bell,
1957) (Table 2). Many of these isolates that grow on 2,4-D as sole carbon and
energy source include species putatively assigned to Alcaligenes (Amy et
al.,1985), including Alcaligenes eutrophus JMP134, which contains the well
characterized 2,4-D plasmid pJP4 (Don and Pemberton, 1981). In addition to

19

.6385 iBEOmoEoEu mm Agata aim «5
:88 bummmmumc 8: 806 mg .mfimbm 9.65 how 3:352 53 as» “o: 96: 5:38wa Dim wfipoucu mEEmmE ..

 

 

was i a .35 «$222 2.8326 RE.
32 :3 a $52 Swao E26 85. 58%..
Aafiv
32 $363— 98 >23» Eatemmcmxru u.»§u2u§§
82 .2 a. 580 233:3
as: .._.w a .820 :8 Scum esz 8538.6 882m 8&8 .m
82 .32.»: c5 §B< 2:23. a 885% a
83 22$: Ea 8~E§< 285 fi "84.5mm .% 8.353%
$2 :1 a 5.35355 . 9 3333.5
83895 was
$2 .3:on use 3.50m . auguMoEcazueutmn Bugiuhm
«ma .coflwtum can :85— . 3.533 .5 Sumtmgu uMfimcSuEam
K2 .Eaau can 3an 3mm“ £28m 98 ESE . in m§§§m
me: .83 . seasz
83 ans—262 use “8:25 . azsmoumi
he 5...”: Ba 8285 $3 56am . £3.52»; m
$2 s33: ea 5:20 9. ma. 2qu 5:53:85
RS few van .6534 . Eatu‘uBuSoU
$2 5: a seeks...“ .._~ a. .83 ssm 13> >52 . .9 taxes:
$253.“; 95 cafism . baggagé
£2 .1 a $8.4. Euro 98 332mm 9 83:8?
$2 defies... 228 £833 95 ER mates: 83:8?
52 £235.: 25 con 2.235 9.8 Ba 333% sexes?
3.2 £9895."— uca con— u=abaa< an manages.“ 853:6;
”81.3.5.8?— £2.22 n58 ER SEE assess regs?
82 escape“... .25 Go 2:53 9:3 ER 3.8:: mafia...» sags?
85$me Ewto mnmm cowaool— Emmaaqu
@355”

 

umfimmzawuo masonic—u unauouamxocoameozumER me an: vats—um .N ozah

20

Gram negative bacteria, several Gram positive 2,4-D degraders have been
reported, including Arthrobacter (Loos et al. 1966); Nocardioides simplex
(Kozyreva et a1. 1993); Streptomyces sp. (Speicher, 1954); and various other
unidentified Gram positive bacteria (Amy et a1. 1985; Tonso et al. in press). In
addition to bacteria, the fungus Phanerochaete chrysosporium has been
shown to degrade 2,4-D, although not as a growth substrate (Yadav and
Reddy, 1993).

In addition to the organisms described above, scores more strains
capable of degrading 2,4-D have been recovered (Ka et a1. 1994; Tonso et al.,
submitted). Because of the great numbers of 2,4-D degraders that have been
isolated, researchers have sought new ways to quickly and easily characterize
these strains to understand their diversity. The diversity of any group of
organisms can be claSsified and organized in two ways: by evolutionary, or
phylogenetic, classification, or by phenotypic classification.

Phenotypic classification: Until relatively recently, phenotypic
characterization has been the only means available for classifying bacteria.
Bergey's Manual of Determinative Bacteriology has been the definitive source
of classification which was based solely on phenotypic classification. Some

techniques for assessing phenotypic characteristics include classical
microbiological techniques (such as the Gram stain), BiologTM, and fatty

acid methyl ester (FAME) analysis (Tonso et a1. 1992). Genetic characteristics
have also been used to assess the diversity of 2,4-D degraders, including

plasmid hybridization (T onso et a1. 1994) and a chromosome fingerprinting
technique using PCR amplification of repetitive extragenomic palindromic

(REP) sequences as primers (Fulthorpe et al. submitted; Ka et a1. 1994; Tonso

21

(REP) sequences as primers (Fulthorpe et al. submitted; Ka et a1. 1994; Tonso et
al. in press).

Although non-molecular techniques, such a classical techniques like
Gram stain and morphology, and newer techniques like FAME and Biolog
provide information on the phenotypic diversity and REP and hybridization
analyses indicate the genetic diversity of these organisms, these methods
cannot be relied on to provide a true phylogeny of these organisms, and thus
the ancestral lineage of these strains. Evidence has shown that phenotypic
groupings do not necessarily reflect phylogenies of these strains. For example,
it was originally thought that all purple, photosynthetic bacteria were closely
related and formed a tight clade to the exclusion of other bacteria until Woese
et a1. (1980) discovered, using comparisons of 16S rRNA and cytochrome
genes, that these purple photosynthetic bacteria were more related to other
Gram negative bacteria than they were to each other. For this reason, I will
use phylogenetic relatedness based on 16S rRNA as one indicator of the
evolutionary relatedness of these strains. Once the ancestry of the strains and
the relatedness of the genes involved in 2,4-D degradation is known, I can
ask whether the diversity of 2,4-D strains arose by divergence or by gene

transfer.

22

Phylogenetic classification: Another way to catalogue the diversity of 2,4-D
degrading strains is using phylogenetic classification using 16S rDNA. This
recently developed technique may fulfill Darwin's prophesy:

"The time will come I believe, though I shall not live to see it,
when we shall have very fairly true genealogical trees of each great
kingdom of nature."

~ Charles Darwin (in Wheelis, 1992)

Studying phylogenies of bacteria has revitalized microbiology
(Margulis, 1992; Olsen et a1. 1994). Unlike eukaryotic phylogenies which can
be based on morphological data, bacterial phylogenies were practically
impossible to do before the advent of molecular sequencing. Molecules used
in phylogenetic analysis are often highly conserved and include cytochromes,
elongation factors, and small and large ribosomal subunits. "Molecular
sequencing technology has brought biology into the era of global (universal)
classification" in which, all organisms can be compared to all others (Wheelis
et a1. 1992). .

Phylogenetic analysis is growing in popularity as molecular techniques
become increasingly widespread. Now that phylogenies based on molecular
sequence data have become better understood, many phenotypic
characteristics, such as biochemical markers, have been discovered to be
phylogenetically relevant. For example, particular fatty acid methyl esters
have been found to be indicators of alpha, beta, and all four gamma
subgroups of Proteobacteria. In addition to these markers, one might ask if
any of the 2,4-D genes are markers for particular phylogenetic groups. This

research will address this question.

23

168 rRNA: Of all the molecules to be sequenced for phylogenetic analysis,
16S rRNA is the most popular, as indicated by the large numbers of sequences
deposited in GenBank and the Ribosomal Database Project and the deeply
rooted, extensive phylogenies produced with this molecule (Figure 5). This
molecule, which is approximately 1500 nucleotides in length in bacteria, is
highly conserved (Figure 6), and is therefore a useful molecule for doing
phylogenetic studies. It is convenient to use this molecule because new
sequences can be compared to large numbers of other 16S rRNA molecules.
Since first sequenced in Escherichia coli (Brosius et a1. 1978), over 2600
additional sequences of small ribosomal sequences have been added to the
rapidly growing Ribosomal Database Project (RDP) and even more are
accessible in Genbank. Comparing small ribosomal subunits has been
extremely useful in creating a universal phylogeny (Woese, 1978; Fox et a1.
1980). In addition to characterizing extant, cultured organisms, 168 rDNA
sequencing has allowed researchers to characterize ancient organisms, such as
Bacilli trapped in bee guts in amber 25-40 million years old (Cano, 1994) and
organisms which have not been cultured (Amann et al. 1994; Schmidt et al.
1991).

24

.Ammmfi in 8 $2855 60.5 .995 uuoswolzm 2833:: 380% .m 339m

3936290
223382:

 

mm‘muosctmcg

 

momcoEono_F
ate—unnosmi

 

33:32“. a m on on?»
~23.» Eatusoim :. o o
2893 «3233.535
6.353: 2.202
9.5 3.2%.“: .235»: -295 520
E 3.8302232
55:. Sommcocmbam 9352
no.2: 5:395:

06.5 «33625 c020

mEEEd

mimoam mmmzocc. mtmcomm

 

 

 

25

100

uoo

 

 

 

 

 

 

 

§

Figure 6. Diagram of 165 rRNA molecule. Heavy lines indicate highly
conserved regions and dashed lines indicate hypervariable regions.

26

Inference of Phylogeny

Determining the diversity of strains based on sequence comparisons
requires the reconstruction of a phylogeny. Reconstructing a true phylogeny
is often a difficult task as illustrated by the fact that the number of possible

unrooted, binary trees for N taxa is

(2N-5)!!

(that is, 1 x 3 x 5 . . . x (ZN-5)) (Penny, 1991). So, for example, 10 taxa would
yield 2,027,025 possible different trees (Penny, 1991). To assist in determining
the correct tree from the huge number of possibilities, several methods canbe
used to infer a phylogeny from sequence data (reviewed by Doolittle, 1990;
Felsenstein, 1988; Goodfellow, 1985; Kumar et a1. 1993; Maddison and
Maddison, 1992; Penny, 1991; and Swofford and Olsen, 1990) The three major
categories of methods, including parsimony, distance matrix and maximum
likelihood, are briefly outlined here.

Parsimony: Parsimony uses the least amount of convergence and reversal
among the characters to determine a tree. The measure of. evolutionary
change is the treelength, which is calculated by the weighted sum of "steps, "
or base pair changes. Programs designed to analyze trees using maximum
parsimony include PAUP, PHYLIP and MacClade. MacClade lets one observe
treelengths and other tree statistics, although it does not automatically
calculate the shortest treelength. PAUP and a program in the PHYLIP package

automatically calculates the most parsimonious trees using a variety of

27

algorithms. Parsimony methods work well if there is little change over
evolutionary time (Felsenstein, 1981; Stewart, 1993). However, parsimony
methods can fail if there are large amounts of change or if some branches of
the tree are much longer than other branches of the tree, since parsimony
tends to group all long branches together, even if they are not related.
Distance Matrix: There are several distance matrix methods which build trees
based on distances between pairs of strains. These techniques include
unweighted pair group method of analysis, UPGMA, the neighbor joining
method and the Pitch and Kitch least squares techniques (Felsenstein, 1993).
The DeSoete Tree Fit package used by the RDP is an example of using the least
squares fitting algorithmon a distance matrix. UPGMA tends to fail if a
molecular clock is not obeyed, that is, if rates of evolution differ between
different taxa being analyzed (Felsenstein, 1981). Another problem with these
methods is that they fail to fully exploit all information in the original data.
Maximum Likelihood: This involves inferring phylogenies from a simple
probabilistic model of evolution (Felsenstein, 1981). Maximum likelihood
"involves finding that evolutionary tree which yields the highest probability
of evolving the observed data" (Felsenstein, 1981, p 369). This is based on a
function of the tree (the hypothesis) and not a function of the data
(sequences). RESTML, which is part of the PHYLIP package, is software
designed to estimate maximum-likelihoods from restriction site data
(Felsenstein, 1992)

Confidence test for phylogenies: Because different methods often yield
different trees, and sometimes more than one optimal tree is found per

method, it is important to estimate the confidence in the tree (Felsenstein,

28

1985). The most widely used statistical tool for this task is the "Bootstrap"
method, which Felsenstein was the first to apply to phylogenies. For this
method to work, many iterations of the tree algorithm are performed, each
one with resampled data. In each iteration of the bootstrap, nucleotides (or
other character types) in a data matrix are resampled such that some
nucleotides (characters) are counted one or more times and other nucleotides
are not counted at all.
Phylogenetic congruency: How reliable is a single molecule, such as 165
rRNA, for determining the phylogeny of an entire organism? This question
has not been extensively investigated and is not yet known how great an
effect gene transfer and gene duplication has had on the evolution of
organisms (Doolittle and Brown, 1994). However, in cases that have been
tested, the deeper branches of the 16S phylogenetic tree have shown to be
congruous with phylogenies derived from using cytochromes, elongation
factors, and ATPases (Iwabe et a1. 1989).
Species Definition: In discussing diversity, the term "species" is often
invoked, although in bacteria this term is difficult to define. There are two
ways to define species: 1) using a phenotypic definition or 2) using a biological
species definition which says members of same species are inbreeding groups.
Adherents of the biological species concept at one extreme claim that
there are no species of bacteria, since there is no genetic isolation among
bacteria (Sonea, 1989). A more moderate view of the biological species is that
of Dykhuizen and Green (1991) who define species as those strains which
demonstrate a high degree of recombination.

The current working definition of bacterial species accepted by most

29

microbiologists is phenetic definition. This definition was agreed to by a
convention of bacterial systematists who decided that two strains that are at
least 70% identical on the level of DNAzDNA hybridization are members of
the same species (Wayne et al. 1987).

Because so many researchers are now using 16S rDNA to classify
bacteria, Stackebrandt and Goebel (1994) have proposed that any two strains
sharing less than 97% similarity of 16S rRNA sequences belong to different
species. Stackebrandt and Goebel (1994) report that "organisms that have less
than 97.0% sequence homology will not reassociate to more than 60%" which
is below the 70% threshold for a species definition. It should be noted that the
divergence of total genomic DNA is not necessarily reflected in the
divergence of 16S rDNA (Table 4). For example, Burkholderia cepacia and
Alcaligenes eutrophus differ by 4.3% at level of 16S rDNA sequences, whereas
these differ by 93% at level of DNA-DNA similarity (Li, 1993).

Our definition and understanding of species may define our
understanding of the diversity of 2,4-D degrading strains. For example, many
2,4-D degrading strains appear to be pseudomonads (Ka et a1. 1994) and like
other pseudomonads may share a high degree of sequence similarity at the
level of 16S rDNA, yet may have diverged considerably at the level of total
genomic DNA (Johnson and Palleroni, 1989). "In contrast, strains of different
species of other genera may show in their DNAs a degree of conservatism
approaching that of the ribosomal genes. It is tempting to attribute these
peculiar properties of the pseudomonads to the ubiquitous occurrence of
these organisms and the concomitant need for adjustment to the selective

pressures of highly variable natural habitats." (Johnson, 1989). Perhaps these

30

 

 

may .._m am .3 $03095. 80mm UUH<
owofimaouflfim was acmcfiom mo H 53332.58 m mmzomomouucu .m

mama .._m «o .5 3va UUH< Rona UUH<
$3. ESE—mm van .3922 5mm 5 Sucks... utufiofitam 2:323» mozumzuuzx

mm

cmmm man
83 385%. mda 3. 383E Samuum untougummncmx Sufism

33:? no .mnmawfipoom w 3.68m Na 8.63m
$3.4m no .988 Wmm av 35:283....“ .2 2532883.. Eztfiogoumg
mom“ :1. no .xom wdo mdm mzkommBSm .m mEEmSfimmA $5.83
8:838. m3 .x. bra—mam u 535 H :36

<ZQ-<ZDo\o

 

cowmfiptéf <ZQ-<ZD mo “condom 95.8., 3353 8:258 m3 9:39:00 38% mo Etna—Em .v 2an

31

pseudomonads adjust to their habitats by genetic exchange.

Determining the diversity of 2,4-D degrading strains could be confused
not only by differences in species concepts, but also by the confusion brought
on by changing species and genus names. For example, Pseudomonas strains
have undergone various name changes in the past 10 years. Strains once
belonging to the genus Pseudomonas have been reclassified into several
groups such as Burkholderia, Sphingomonas, Comamonas, and
Xanthomonas (de Ley et a1. 1986; Stackebrandt et a1. 1988; Yabuuchi et al.,
1992). In order to help sort out changing names and classifications, Table 3
shows past and present names and classifications of selected strains of

Proteobacteria.

32

 

 

 

 

 

 

 

 

 

3.3.2.2.. 3%....33. 3.2%.... 32.83»... n 9980
. , a 9580

83.55.. gage». 33.95.. 3%.. a 9:5

n 80D

3.38...» 853.3? 2:38...» 85M..8.< N San
362.8 a...u2o§§m sauna negoEonauum N Bun
3238.8. mucoEquU 3838.8. «38.5.38.» — Bum
sage... , .2.s&..2..lmfl§ss§& sass... a . J is:
£3... :...§..e§...§. is... A §£..:§§ . _. : a .. 2.2
32.3.... .§§ug...mé. ., sass... 3%.... . :12
8:3: @333— 8.5.2 20 R320 awe—98$ 95.0: 88335
<2... «395838.. 53 .383 Ba. .8025

.._:o.o=um

 

.3383 29?: Shtfiosem mam—o no. .3...— ougfioaofioz .m «in...

33

DIVERSITY OF 2,4-D DEGRADING GENES

In the previous section I discussed methods and issues for assessing the
diversity of bacteria. However, this knowledge reveals nothing of a strain's
genetic constitution for degrading 2,4-D. In this section I will discuss what is
known about the diversity of genes for degrading 2,4-D and related
chloroaromatic compounds.

Gene Similarities: There has been little comparison of the genes responsible
for degrading 2,4-D. The entire pathway, tfdA-tde, has been sequenced only
in IMP134(pIP4) (Don et a1. 1985; Perkins et a1. 1990, Streber et a1. 1987). The
tfdA gene has subsequently been sequenced and found to be identical in
strains EML159 and TFD38 (Fukumori, personal communication) and
pCSV90 (Bhat et al. 1994) and found to differ by 23 percent in Burkholderia
strain RASC (Suwa et a1. 1994) and Burkholderia strain TFD6 (Matheson,
1994), the latter two of which are identical to each other.

It appears that alternate pathways and / or other genes involved in 2,4-D
degradation differ substantially from those genes in the pathway of pIP4. This
is suggested by 2,4-D organisms which do not hybridize to pJP4 in colony
hybridization experiments. For example, Matheson et al., (1991) assessed the
diversity of 2,4-D strains from different sources and found that pJP4
hybridized to only 16 of 52 (30%) of the 2,4-D degrading colonies isolated from
three different locations. Furthermore, Holben et al., (1990) found that total
bacterial DNA isolated from a Saskatchewan soil exposed to 2,4-D hybridized
with genes tfdA and tde, but had no significant hybridization to tde, D, E, or
F indicating a population of 2,4—D degraders different than that of pJP4. Ka et

34

a1. (1994) isolated 2,4-D degrading strains from a Michigan agricultural plot,
some of which contain plasmids that hybridize to tfdA only, others to
tfdA,B,C,D and still others which had no apparent similarity to pIP4.

Other chloroaromatic degradation genes: Other studies have examined the
diversity of genes related to tde, a catechol dioxygenase, from other
chloroaromatic degradative pathways such as the 3-chlorobenzoate (3C3)
pathway and trichlorobenzene (TCB) pathways (for review see van der Meer,
1992 ; Harayama and Kok, 1992; and Fukuda, 1993; Ghosal and You, 1988;
Ghosal et al. 1985). These studies showed that gene analogs of tde, D, E, F
were in the same orders in operons found in the 2,4-D pathway of JMP134, in
the 3C8 pathway in plasmid pAC127, and in the TCB pathway of
Pseudomonas P51 (Figure 7). Gene similarities can be determined by amino
acid similarity, DNA sequence similarity or by gene probing. In contrast,
hybridization analysis with various probes made from 2,4-D genes and 3-
chlorobenzoate (3C8) genes (which have sequence similarity to the lower
pathway of 2,4-D degradation in pIP4) offers a means of more efficiently
surveying sequence similarity of the genes for 2,4-D degradation.

35

Ga. :3 s .82 8.. §> 88..
$932.68 £3898 95...? .o .3352. «wage—35.8 .8. 35w .0 cofluamfiou N 8:»...

 

 

 

 

 

 

 

 

 

 

 

 

3085020 EU flag . H LL 5.38
«Scuba SB :mummo
a: M@ DD a.
s: m m G

c...

Roi. A

«3.2

U Q to .. ... 223.3%

m S... k
co_.m~__muo_ 28:00 I In
. n.3-

 

sac.

36

GENE TRANSFER POTENTIAL

To understand the role of gene transfer in the evolution of 2,4-D
degrading strains, it is useful to review some of the literature pertaining to
gene transfer mechanisms in general, what is known about mobile 2,4-D
degrading genetic elements in particular, and about gene transfer in nature.

Although conjugal gene transfer among certain types of 2,4—D degraders
is documented (Don and Pemberton, 1981), it is not known to what extent
gene transfer is responsible for the diversity of 2,4-D degraders observed. One
aim of this research was to assess the ability of various 2,4-D degrading strains
to transfer the 2,4-D phenotype. In addition, this project examines other
modes of gene transfer aside from conjugation. Conjugation has been the
only means of gene transfer reported for transfer of the 2,4-D phenotype.
Horizontal Gene Transfer: Interspecies horizontal gene transfer has been
observed in a huge variety of strains, crossing all sorts of phylogenetic
boundaries including species,lgenus, order, and even kingdom (Figure 8) (for
review see Heinemann, 1991 ; Sikorski et a1. 1990; Stachel and Zambryski,
1989). Transfer has been shown to occur by transformation (Frischer et a1.
' 1990) transduction, conjugation of plasmids and transposons on these
plasmids (Fulthorpe and Wyndham, 1991 ; Nakatsu, 1992) and retro-transfer
(Top et a1. 1994).

Perhaps mobile genetic elements are as important, or more important,
than the chromosome itself in determining the persistence of a catabolic
phenotype, particularly in pollution stressed environments. Different

bacterial hosts may or may not survive in a stressed environment, but mobile

37

\-

 

Figure 8. Diagram of interspecies gene transfer.

38

genetic elements might survive if transferred to an appropriate indigenous
host. Populations of non-native organisms often are reduced to below
detectable levels When introduced into an environment (Steffan et al.1989;
Jain et al.,1987; McClure et al.,1989) thus minimizing their effectiveness in
bioremediation. Perhaps the effectiveness of these introduced strains would
be increased if the degradative genes were on mobile genetic elements.
Conjugation: Since the genes for 2,4-D degradation, like many other catabolic
pathways, are often plasmid encoded, conjugation may play a major role in
dispersing and evolving new catabolic pathways in the environment
(Heinemann, 1991; Berg and Howe, 1990). For example, it has been
hypothesized that "module's" of distinct parts of the 2,4-D pathway of pJP4
have been sequentially assembled on a plasmid (Streber et al.,1987).

Over 100 plasmids capable of catabolizing recalcitrant and manmade
organic compounds have been isolated (Chakrabarty and Gunsalas, 1971;
Mergeay et al., 1990; Sayler et al., 1990; Tirnmis and Piihler, 1979). Plasmids
may be important for distributing genes involved in the degradation of
pollutant compounds, (Sayler et al., 1990; Mergeay et al., 1990). It has been
reported that when there is a high concentration of a contaminant, for which
degradative genes are encoded on plasmids, the frequency of these plasmids
increases in the population. For example, fewer than 5% of bacteria in
pristine aquifers have been shown to contain plasmids, yet nearly 40% of the
bacteria in contaminated aquifers contain plasmids (W ickham and Atlas,
1988).

Conjugation is a mode of horizontal gene transfer that allows exchange

of genetic material within species, between species, between genera (Brisson-

39

Noel et al.,1988; Chatterjee and Chatterjee, 1987; Clewell, 1993; Datta and
Hedges, 1972; Hardy, 1987; Mazodier et al., 1989; Natarajan and Oriel, 1991 ;
Thomas, 1989; Trieu-Cout et al.,1987; Willets, 1985) and even between
kingdoms (Heinemann and Sprague, 1989). Several plasmid incompatibility
groups, including IncP1 (e. g. RP4), Ian (e. g. RSFlOlO), IncN, and IncW can
transfer to a wide range of species at relatively high frequencies (Bagdasarian
et al., 1981; Frey and Bagdasarian, 1989; Simon et al,1981; Thomas, 1989). The
features of these broad host range plasmids that allow them to be transferred
to, and maintained in, a broad range of organisms are that they contain
relatively few restriction sites and can replicate in a wide range of hosts by
having special, largely self-contained replication machinery. Interspecies
gene transfer is limited to some degree by the ability of plasmids to be
transferred from one species to another, but is limited more by the ability of
that plasmid to be maintained in various hosts (Sayler et al., 1990; Heineman,
1991).

Plasmid pJP4: The best studied plasmid for 2,4-D degradation, plasmid pJP4
(Don and Pemberton, 1981; Figure 9) is a broad host range, self transmissible
plasmid of the IncPIB incompatibility group. The 87.9 kb pIP4 has been
shown to transfer freely to E.coli, Rhodopseudomonas sphaeroides,
Rhizobium sp., Agrobacterium tumefaciens, Pseudomonas putida,

Pseudomonas fluorescens, and Acinetobacter calcoaceticus at frequencies
ranging from 10'1 to 10'7 (Don and Pemberton, 1981; Friedrich et al., 1985).

Although pJP4 can be transferred to different species and genera of bacteria,
the phenotype for 2,4-D degradation is observed only in A. eutrophus,

4O

EcoR1 0.00

   
  
 
   
 
    
 

EcoR1 80.00

ECOR1 67.00

EcoR1 66.00
BamHl 65.50

EOOR1 65.00
-amH| 28.00

EcoR1 30.00
amHl 33.00

 

Boom 53.00
BamHl 52.00
EcoR1 43.00
BamHl 47.00 Baml-ll 43,50
EcoR1 44.00

Figure 9. Diagram of plasmid pJP4 from Alcaligenes
eutrophus IMP134.

41

A. paradoxus, P. cepacia and P. putida (Table 5). This phenotype is not
detected in pJP4 transconjugants of Agrobacterium, Rhizobium, E. coli,
Rhodopseudomonas, Acinetobacter, P. fluorescens and P. aeruginosa. One
possible explanation for the limited host range of expression of genes for 2,4-D
degradation is that the final gene in the pathway, maleylacetate reductase
(mar) is Chromosomally encoded, such that only strains with this gene can
metabolize 2,4-D as their sole source of energy (Kukor et al.,1989).

Other 2,4-D plasmids: Other 2,4-D degraders have been found with 2,4-D
genes located on plasmids (Amy et al.,1985; Chaudry and Huang, 1988; Don
and Pemberton, 1981; Ausmees and Heinaru, 1989; Bhat et al. 1994).
Pemberton et al., (1979), found that of 22 colonies isolated on 2,4-D, 17

produced 2,4-D negative segregants (tfd') after only 15 to 20 generations when

cultured in non-selective broth, suggesting loss of a plasmid encoding genes
for 2,4-D degradation. The remaining five strains all contained plasmids
conferring ability to catabolize 2,4-D. Three of these were similar to pIP4 in
size and restriction pattern and incompatibility group, whereas pIP2 and pIP9
differed in size, incompatibility group and had the ability to degrade
phenoxyacetic acid (when not induced by 2,4-D), which pJP4 does not.
Chaudry and Huang (1988) found another plasmid, pRClO, encoding a 2,4-D
pathway in Plavobacterium. This 45 kb plasmid has sequences that hybridize
to tfdA and tde of pIP4, but has no sequences that hybridize to the regions of
pJP4 involved in conjugation, incompatibility, and maintenance. Amy and
colleagues (1985) found plasmid pEML159 in Oregon sludge which was

similar to pIP4 in size, restriction endonuclease fragment patterns, and ability

Table 5. Recipients of plasmid pJP4 from Alcaligenes eutrophus JMP134
indicating ability of recipients to express 2,4-D phenotype.

42

 

 

Proteobacteria Frequency Growth
Recipients Subdivision of transfer on 2,4-D Ref‘
Alcaligenes eutrophus Beta 10-50°/o + 1
Variovorax paradoxus Beta 10-50°/o + 1
Burkholderia cepacia Beta 50% + 2, 3
Pseudomonas oxalaticus nd + 2
Pseudomonas sp. PK01 + 4
Burkholderia pickettii Beta + 5
Burkholderia putida Gamma nd - 2
Burkholderia putida ‘ Gamma nd + 1
Escherichia coli Gamma 4 x 10‘4 - 1
Pseudomonas fluorescens Gamma 7 x 10‘4 - 1
Alcaligenes calcoaceticus 1 x 10‘3 - 1
Pseudomonas aeruginosa Gamma - 2
Rhodopseudomonas Alpha 5 x 10'6 - 1
sphaeroides
Rhizobium Alpha 1 x 10'5 - 1
Agrobacterium Alpha 2 x 10'2 - 1

 

* References: 1: Don and Pemberton, 1981 ; 2=Harker,et al., 1989; 3=this

lab;4=Kukor, et al, 1989; 5 = Friederich, et al., 1983.

43

to transfer the 2,4-D degrading phenotype to other bacteria. This similarity is
perhaps surprising since pJP4 was isolated on the other side of the world in
Australia. Ausmees and Heinaru (1989) found three other 2,4-D degrading
strains in Estonia which contain plasmids for 2,4—D degradation.

In addition to pJP4, pRClO, pEML159 and the Estonian plasmids, many
other plasmids have been found in 2,4-D degrading strains (Tonso et a1. 1994
and Ka et a1. 1994). It is still not known whether the 2,4-D degradative
pathways in these strains are plasmid borne but presumably some of these
plasmids are responsible for the 2,4-D phenotype since they hybridize to pIP4.
Others plasmids which have no sequence similarity to pIP4 may or may not
be responsible for the 2,4-D phenotype.

Some 2,4-D strains appear to be on plasmids which are capable of
integrating into and excisizing from the chromosome. Ka and Tiedje showed
that strain P. pickettii strain 2118 had variants with 2,4-D genes alternating
between the plasmid and chromosome, and that under further selection for
2,4-D growth, a second, smaller version of the plasmid was excised which
apparently left tfdA behind on the chromosome.

Chromosomally encoded 2,4-D genes: Although the best characterized genes
for 2,4-D degradation are plasmid encoded (Fisher et a1. 1978; Don and
Pemberton, 1982; Friedrich et a1. 1983; Chaudry and Huang, 1988; Ausmees et
a1. 1990; Amy et al. 1985; Bhat et al. 1994) there is evidence that many other
2,4—D degraders have Chromosomally encoded 2,4-D genes, including RASC
(chapter 4; Suwa, in prep); TFD6 (Matheson,et al., 1994) and BRI6001 (Greer et
al. 1990) which appears to be identical to TFD6. Ka et al. 1994 also found that

genes for 2,4-D degradation were Chromosomally located. It is not known

44

whether chromosomally located 2,4-D genes can be transferred.

Gene transfer in the environment: The first in situ observation of plasmid
transfer in nature involved antibiotic resistant plasmids and introduced
strains of E. coli as donor and recipient in sterile soil (Weinberg and Stotzky,
1972). Since then, many researchers have investigated plasmid transfer in
nature and several review articles and books have been written on this topic
(Levy and Marshall, 1988; Coughter and Stewart, 1989; Trevors et al.,1986;
Stotzky,1989; Stotzky and Babich, 1986; Levy and Miller, 1989; Fry and Day,
1989). Most of these studies have focused on transfer of antibiotic resistance
between strains of introduced bacteria. For the sake of brevity, I will discuss
only selected studies, especially those which focus on transfer of catabolic
plasmids and transfer to autochthonous soil microorganisms (Table 6).

To demonstrate that soils contained potential indigenous recipients of
plasmids, Schlif and Klingmiiller (1983) showed that 1.3% of 600 randomly
isolated soil bacteria could be recipients of pRD1, a plasmid encoding
resistance to carbenicillin, kanamycin and tetracycline, from E. coli.

In live soil, conjugal plasmid transfer has been demonstrated between
introduced strains. Trevors and Oddie (1986) demonstrated transfer of
tetracycline resistance of the R plasmid between strains of introduced E. coli
in soil. Transfer of heavy metal resistance plasmids from E. coli to
Alcaligenes eutrophus has been demonstrated in live soil amended with
nutrients (Top et al., 1990). In addition, plasmid transfer has been shown to
occur in soil between introduce strains of Bacillus (van Elsas et al., 1987) and

between introduced strains of Streptomyces (Wellington et al., 1990).

45

 

 

9.... 3.2.... 8&2. 3.3....» .< Sesame... =8 .1 5 day .3.
ion as"... .83...... 2.2 33.88.... .< 55.8 :8 .258 9:. A3.3.2.. .8.
2...; 3.... ch 895.65 3......“ .m =8 3... .86 Ba 5.222 .8.

8.. 80...... 38.6... 3.... .3 eggs? 5888533522.? as. 2.853". .8.
”10.528. 33.... 36...? =8 .m 5.5.8 :8 6.86m .25 3:85.. 8...
mm-.. 33.032... 63: 3...... .m 9.3.5. 3...... a 5888.5 8.5... c.2932 8....

«:9. «is... 3&6... =8 .m =8 25 625.8 .3...W 9285: 8....
managesps .m 8.39... unificatfi =ow .3 3 £995.35 83

«as. do. 2038955838.. 3.598. 3...... .m 6893.... 8.32m .1 a .8282 82
a»: 36.6... 3.3.5. 3...... .m .3833. 85...... as... .3 9:. E .23 8...
3533 Exm nan £333....m Eataggawu. .M :3 03.— ..a .9 £85.. 83

CNN—hum was .m marzazm .m =0» w—tmum mgr—u m>m..— AG “0 inn—m Ca.) hwmw

mam 5...; 268... m «C. =8 .m 39.2 =8 .m =8 25 22.0 can .295; 8....
man «man 390»? 3:23....qu 5.5.8 :om an 3 3.638.. 33
53.830 Ex 8.28. 3.8%.... =8 .m e... g. 3.58%.... as. :28. 82
89.8.2. 3......“ .m :8. .8... .95 85.20 m3.

.+w.m $03 m :8 .m :8 .m =8 «imam 5.50% 9.3 93:83 N53
buy—an: Vagina—h «GomthOM— H88 3?: Faun—8.08"— Gunfi—

 

duaa: a. 09.3.86 2.90 .c wink

46

The few studies demonstrating transfer to indigenous soil bacteria,
generally have involved transfer of antibiotic resistance. For example,
Henschke and Schmidt, (1990) demonstrated the transfer of the RSFlOlO-
derived, kanamycin resistant recombinant plasmid pFL119-2, from E. coli into
indigenous soil microorganisms. In their system, the kanamycin resistance
gene was cloned behind a strong, broad host range tac promotor and in front
of a strong terminator. The plasmid was mobilized by the transfer functions
of RP4 cloned into the chromosome of the donor strain, E. coli, Although
not necessarily resident on a plasmid, tetracycline resistance from the
conjugative transposon Tn916 has been shown to transfer from Bacillus
subtilis into indigenous soil actinomycetes (Natarajan and Oriel, 1991).

Transfer of catabolic plasmids has been demonstrated in various
microcosm systems. For example, plasmid pWWO-EB62, a recombinant TOL
plasmid with the ability to degrade ethylbenzoate, was shown to transfer from
an introduced strain of P. putida KT2440 carrying the plasmid into the
introduced recipient P. putida UWCl in an activated sludge microcosm
(NiiBlein et al., submitted). Transfer of the catabolic plasmid pBRC60,
encoding 3-chlorobenzoate (3C8) degradation ability, into indigenous
microorganisms has been demonstrated in a freshwater mesocosm
(Fulthorpe and Wyndham, 1991). McClure et al., (1991) demonstrated
transfer of the non self-conjugative 3C8 recombinant plasmid, leO, from P.
putida UWC5.10, an indigenous strain with transfer functions, to 19 different
morphotypic, gram negative, oxidase positive strains in an activated sludge
bioreactor.

Transfer of catabolic genes to indigenous soil microorganisms has been

47
demonstrated in two cases in irrigated soil columns. Pertsova et al., (1984),
inoculated soil with 3.9 x 107 P. putida with a plasmid encoding enzymes for

3-chlorobenzoate degradation. A stream of mineral medium with 0.1% 3CB
was run through 30 g of soil in column at a rate of 40-50 ml per day for 10 or
30 days. They found that the 3CB phenotype had transferred into strains in
the inoculated soil column, but not in the control. Although the strains
differed in various phenotypes from the introduced strain, no molecular or
genetic analyses were employed to determine whether these were different
strains or simply mutants of the original donor. Brokamp and Schmidt (1991)
used a 40% soil/ 60% sand column to demonstrate the transfer of plasmid
pFL40 (carrying genes for Dalapon (2,2-dichloropropionic acid) degradation)
from Alcaligenes xylosoxidans ABIV into indigenous microorganisms. After
50 days in a soil column inoculated with the donor, various indigenous soil
microorganisms were isolated which had received and expressed the plasmid,
including a Pseudomonas species, P. fluorescens, P. paucimobilis, and

Alcaligenes sp.

SUMMARY
In this chapter I have provided the background needed for assessing
the role of gene transfer in the evolution of 2,4-D degrading strains. Three of
these issues are: 1) the diversity of the strains; 2) the diversity of the genes
involved in 2,4-D degradation; and 3) the transfer potential of these genes.
Here I have presented information on the diversity of bacteria capable

of degrading 2,4-D and discussed various methods which have been used to

48

analysis for evaluating the diversity of bacteria in general. Because it
provides unambiguous data, and because a large database exists for sequence
comparison, 165 rRNA sequence analysis is the method I have chosen to
measure diversity of 2,4-D degrading strains.

The second issue to be addressed in understanding gene transfer in
evolution of 2,4-D degraders is determining the diversity of 2,4-D degrading
genes within these strains. In this chapter I have shown that one
2,4-D pathway has been extensively studied, yet other pathways known to
exist have been only partially characterized or still remain unknown. Also,
the distribution of the known pathways has been poorly studied. One way to
examine the distribution of tfd genes is to hybridize the tfd genes of the well
characterized plasmid pJP4 to various strains. This has been done by Ka et a1.
(1994). However, they used stringency hybridizations which would not detect
tfd genes less than 90% identical to those of pJP4. In chapter 2,1 describe low
stringency hybridization experiments done collaboratively with Dr. Roberta
Fulthorpe to detect additional, related 2,4-D degradation genes. In addition to
hybridization analysis, the tfdA gene was sequenced to determine additional
diversity of this gene.

The third issue in better understanding the role of gene exchange in
generating the divert of 2,4-D degraders, is the potential for gene transfer
potential. In the last part of this chapter I reviewed a small portion of the vast
literature describing gene exchange as it occurs in the laboratory and where it
has been shown in nature. In the last two chapters of this dissertation, I
describe experiments investigating a somewhat unusual exchange of 2,4-D

degrading genes in laboratory studies.

49

degrading genes in laboratory studies.
Taken together, these studies conclusively show that gene exchange
has played an impdrtant role in the formation of the 2,4-D pathways found in

2,4-D degrading populations in nature.

Chapter Two

Phylogeny of 165 rDNA Genes from Diverse
2,4-Dichlorophenoxyacetate Degrading Bacteria

50

Phylogeny of 168 Ribosomal DNA from Diverse
2,4-Dichlorophenoxyacetate Degrading Bacteria

SUMMARY

Many 2,4-D degrading bacteria isolated from various sites in North
America and Australia have been shown to be phenotypically and
genotypically diverse, yet the phylogenetic relationship of these organisms
was previously not known. The phylogenetic relationship of these strains
was determined using two methods of 16S ribosomal DNA analysis. One
method compared partially sequenced 168 genes and the other compared 168
ribosomal DNA using restriction fragment length polymorphisms (RFLPs)
from PCR-amplified 168 rDNA. These fragments were digested with four-
base cutting restriction enzymes and the resulting RFLP patterns were
subjected to phylogenetic analysis. These strains were found to be
phylogenetically distributed throughout the alpha, beta and gamma
subgroups of the Proteobacteria. Of the 23 2,4-D degrading strains examined,
17 had unique 16S rDNA sequences, differing by up to 28 percent of their
nucleotides. The majority of these strains were members of the beta subgroup
of Proteobacteria and six of these were most closely related to the genus
Burkholderia. Three 16$ rDNA sequences were closely related to members of
the beta subgroup of Proteobacteria most closely related to Rhodoferax
fermentens. One salt tolerant strain was found to be closely related to the
genus Halomonas in the gamma Proteobacteria. Five were members of the
alpha Proteobacteria related to the genus Sphingomonas; and one was found
to be most closely related to Rhodopseudomonas palustris from the alpha

subgroup of Proteobacteria.

51

52

INTRODUCTION

"Nothing in biology makes sense except in the light of evolution."

- Theodosius Dobzhansky, 1973

This observation is particularly true for studying gene transfer since it
is difficult to study gene flow without knowing how strains are related. This
first part of the dissertation examines the phylogenetic relationships between
2,4-D degrading bacteria using 168 rDNA as a molecular marker.

The pesticide 2,4-D is quickly degraded in many environments due to a
broad diversity of organisms capable of using 2,4-D (Alexander et al., 1969;
Tiedje et al., 1969; Amy et al., 1985; Ka et al., 1993; Matheson et al., 1991;
Vallaeys, personal communication; Mae et al., 1993). Although these strains
have tentatively been assigned to include species in the genera of
Arthrobacter (Loos, et al, 1966; Tiedje et al., 1969) Plavobacterium,
Streptomyces (Speicher, 1954), Alcaligenes (Bounds and Colmer, 1965), and
Pseudomonas (Evans and Smith, 1954; Greer‘et al., 1990) the evolutionary
relatedness of these strains remains in some cases uncertain due to
ambiguities in bacterial nomenclature. For example, the genus designation
Pseudomonas (despite its name meaning "false unit") continues to cause
much confusion, since this " genus" comprises member species which are
scattered throughout three of the four classes of the bacterial subdivision
known as Proteobacteria (Woese, et al., 1984;Woese, 1994).

The diversity of 2,4-D degraders has been assessed by a variety of

means, including classical bacteriological methods (Alexander et al., 1969;

53

Amy et al., 1987); by substrate utilization patterns using BiologTM; fatty acid

methyl ester (FAME) (Ka and Tiedje, 1994; Tonso et al., 1992), plasmid
hybridization (Matheson et al., 1991) and by genomic fingerprinting using
PCR amplification of Repetitive Extragenomic Palindromic (REP) analyses
(Ka, 1994; Tonso, et al., 1994; Fulthorpe, in prep) and by polyphasic taxonomic
methods incorporating all of the above methods (Tonso et al., 1994). Classical
methods, FAME, and Biolog provide information on the phenotypic
diversity, and REP and hybridization analyses indicates the genetic diversity
of these organisms, but none of these methods provides a rigorous phylogeny
of these organisms. In contrast, it is hypothesized that 16S rDNA sequence
analysis provides reliable bacterial phylogenies which in most cases have
been confirmed by phylogenies derived from other sequence data such as
cytochromes and elongation factors (Olsen et al., 1994; Woese, 1987). For this
reason, comparative analyses of 16S rDNA sequences was used to assess the
evolutionary relatedness of these strains.

In this study, we looked at the diversity of 23 2,4-D degrading strains
isolated from various parts of the North America and Australia by two
methods involving 16S rDNA. One method compared RFLP patterns
derived from 16S rDNA digested with various 4-base cutting restriction
enzymes, a method referred to as Amplified Ribosomal DNA Restriction
Analysis (ARDRA) (Massol-Deya, 1994). Restriction pattern analysis of 168
genes has been successfully employed as a rapid system to analyze
populations of bacteria from bioreactors degrading toluene (Massol-Deya,
1994); in estimating the diversity of strains from hydrothermal vent microbial
mats (Moyer et al., 1994); in identifying various strains of Rhizobia (Laguerre

54

et al., 1994) and symbionts of Flashlight fish (Wolfe and Haywood, 1991); and
for sorting clones before sequencing (Giovanonni et al., 1991).

The second method used in analyzing these 2,4-D degrading strains
involved comparing partial 16S rRNA sequences (Woese, 1981). A
commonly used 168 primer, 519R, was used to obtain sequences from
positions (corresponding to E. coli) of approximately 500 to 200. These
sequences were compared to sequences in the Genbank database. Sequencing
of 16S rDNA has been employed to infer the phylogenies of hundreds of
strains including more than 2600 strains in the Ribosomal Database Project
(Olsen et al., 1993, 1994), and for many smaller projects (e. g. Dewhirst et al.,
1989; Giovanonni et al., 1990; Hiraishi et al., 1991; Li, et al., 1993; Olsen et al.,
1993; Pace et al., 1986; Ward et al., 1993; Yabuuchi et al., 1992). More recently,
several 16S rDNA sequencing studies have analyzed bacteria capable of
degrading various xenobiotic compounds including toluene (Zhou et al.,
1994); polychlorinated biphenyls (Tonso, et al., 1994; Williams, 1994) and

other xenobiotic compounds (Busse et al., 1992).

55

MATERIALS AND METHODS

Media and growth conditions: Cultures were grown in MMO medium
containing 500 or 250 ppm 2,4-D (Sigma) in a minimal mineral salts medium
(Stanier, 1966 ). All cultures were grown at 30°C. Broth cultures were aerated
by shaking on an Innova 4300 shaker (New Brunswick Scientific Co., New
Brunswick, NJ) at 200 rpm.

Bacterial strains: Strains used in this study are listed in Table 1. Strains RASC
and EML 159 were isolated from return activated sludge in Corvalis, Oregon,
and EML 146 was isolated from the Willamette River in Corvalis, Oregon
(Amy et al., 1985). Alcaligenes eutrophus IMPl34(pIP4) (Don and
Pemberton, 1981) was provided by Dr. R. H. Olson. Strains with prefix TFD
were isolated by Holben et al., from the Kellogg Biological Station soils,
Hickory Corners, Michigan, and from the Experimental Farm soils, Indian
Head, Saskatchewan, Canada (Tonso et al., 1995; Holben et al., 1992).
Pseudomonas cepacia strain BRI6001 was isolated from peat in Nova Scotia,
Canada (Greer et al., 1990). Strain E2w1, isolated from soil in the El Loquillo
forest of Puerto Rico, and those strains with prefix B6 (from Quebec) were
from Dr. Roberta Fulthorpe. Strain 1443 and K712 were isolated from the
Geneflow soil plots at the Kellogg Biological Station, Hickory Corners,
Michigan (Ka et al., 1994). Strain M1 is an oligotrophic 2,4-D degrader isolated
by Dr. Roberta Fulthorpe from the island of Moorea, French Polynesia. Strain
2,4-D1 was isolated by Robert Sanford from a herbicide wastewater facility in
Montana.

2,4-D Biodegradation: Disappearance of 2,4-D from the culture medium was

56

monitored by HPLC. The 2,4-D medium with cells was centrifuged at 14,000
rpm in a microfuge and filtered through 0.2 u filters. These samples were
then analyzed by HPLC on Lichrosorb Rp-18 column (Anspec Co., Ann Arbor)
using 60% methanol and 40% 0.1% H3PO4 as the eluant. 2,4-D was detected by
measuring light absorption at 230 nm.

Amplification of 168 genes: Prior to amplification by PCR, DNA was isolated
using the DNA extraction procedure of Ausubel (1992). DNA from strains was
subjected to PCR using primers to amplify most of the 16S rDNA of
Eubacteria. These primers included fDl (E. coli position 8-27F = 5'AGA GT1"
TGA TCC TGG CTC AG3') and rD1 (E. coli positions 1541-1524 = 5'-AAG GAG
GTG ATC CAG CC A-3') (Weisburg et al., 1991) synthesized by the
Macromolecular Sequencing Facility at Michigan State University. The PCR
reaction mixtures were prepared according to the manufacturer's protocol
(Perkins Elmer, Cetus). A master solution was prepared containing the
following: 77.5 [11 ddeO; 10 ul PCR Buffer; 2 pl each of 2 mM dCTP, dGTP,
dATP and d'I'I'P; 1 pl of 20 pmol/ul primer rD1; 1 ul of 20 pmol/ul primer
fDl; 0.5 ul of 5U/ul Taq polymerase (Perkin Elmer, Cetus). Thermal cycling
was done in a Perkin Elmer 9700 Thermal Cycler using the following

conditions: melting at 92°C for 1 min; cycling 35 times at 92°C for 1 min, 10 s;

55°C for 30 s; 72°C for 2 min 10 s followed by a final extension of 6 min 10 s at
72°C.

Restriction Fragment Length Polymorphism Analysis: DNA was digested
according to manufacturer's directions using restriction enzymes from
Boehringer Mannheim and Gibco BRL (Gaithersburg, MD). The enzymes
used were Hue III, RsaI, AluI, SauAl and Msp. Digested products were

57

separated by electrophoresis on a 1.75% agarose gel (Boehringer Mannheim)
at 38V for 12 hours. DNA molecular weight markers included bacteriophage
lambda digested with HindIII (Boehringer Mannheim) and a 100 bp marker
(Bethesda Research Laboratories, Gaithersburg, MD). Gels were stained with
ethidium bromide and photographed with Polaroid positive negative film 55.
The presence or absence of restriction fragments was entered into MacClade's
data editor (Maddison and Maddison, 1992) and this data was used for the
program RestML, a restriction fragment analysis program written by
Felsenstein to determine a phylogenetic tree based on a maximum likelihood
method (Felsenstein, 1992).

168 rDNA Partial Sequencing: PCR products were purified using the Gene
Clean Kit (Bio 101, La Jolla, CA). Purified DNA was eluted from glass beads in
50 ul H20 and quantitated using gel electrophoresis by comparison with a

DNA mass ladder (Gibco, BRL, Gaithersburg, MD). Approximately 150 [.11
DNA template was added to 25 pmols primer and brought up to a volume of
20 111 with H20. The sequencing primer used was 519R (5' GTA TTA CCG
CGG CT G CT G G -3') synthesized by the Macromolecular Sequencing Facility
at Michigan State University. This primer was chosen because it is a
eubacterial version of one of the universal primers which has been shown to
successfully amplify partial 168 DNA from every organism tested (Lane et al.,
1985) and because many researchers have previously used this probe and thus
there are a larger number of sequences in the database (Dewhirst,et al., 1992;
Hiraishi et al., 1992; Lane et al., 1985; Paster et al., 1994; Schmidt et al., 1991).
Sequencing, using fluorescently-labeled dye termination, was performed at

Michigan State University Sequencing Facility using the Applied Biosystems

58

Model 373A automatic sequencer (Perkin Elmer Cetus).

Sequence Analysis: Sequences from 506 to 206 (according to E. coli
numbering) were compared to sequences in databases from Genbank.
Comparisons were made using the Basic Local Alignment Search Tool
(BLAST) from the National Center for Biotechnology Information (Altschul,
1990). Those strains with highest maximal segment pair (MSP) scores were
retrieved from Genbank (Table 2 lists accession numbers and ATCC
designations) for building a phylogenetic tree. Sequences were aligned
manually using the software Squd (Applied Biosystems) and with MacClade
(Maddison and Maddison, 1992). A distance matrix was compiled based on
the number of nucleotide changes per hundred base pairs using software
programmed in HyperCard and with the Dnadist program from the PHYLIP
package version 3.5 (Felsenstein, 1993) using the Jukes-Cantor model of base
changes. Ambiguous nucleotides were deleted from the sequence alignments
leaving 283 unambiguous sites before preparing the distance matrix.
Phylogenetic trees were constructed using the neighbor joining distance
matrix method (Felsenstein, 1993) and the maximum parsimony method

using PAUP (Swofford, 1991).

59

RESULTS

Twenty three strains of 2,4-D degrading isolates and one non-2,4-D
degrading strain of Pseudomonas cepacia strain D5 (aka Burkholderia
cepacia) were selected (Table 1) from a larger collection of 2,4-D degraders on
the basis of their uniqueness based on REP-PCR (Fulthorpe, personal
communication), polyphasic characterization (Tonso et al., submitted) and
hybridization patterns (next chapter). Two pairs of these strains, TFD41 and
TFD38, and TFD6 and TFD20, appeared to be identical by all criteria examined.
These four strains were isolated from Michigan samples. These were
included to determine whether they contained identical 16S rDNA
sequences.
Analysis of restriction patterns from 165 rDNA genes: After PCR
amplification, DNA from individual strains was digested with each of five
different restriction enzymes which recognize different four bp sites. Products
were separated on a 1.75% agarose gel which allowed fragments as small as 50
nucleotides in size to be resolved. Of the 23 strains analyzed by this method,
15 had unique restriction patterns when digested with Msp I (Figure 1a), 13
when digested with HaeIII (Figure 1b); six when digested with Sau3A (Figure
1c); 10 with 111111 (Figure 1d) and 11 types when cut with RsaI (Figure 1e). In
the latter three gels, E. coli was added to serve as an outgroup in phylogenetic
analyses.

Several strains had identical 16S RFLP patterns, including TFD41 and
TFD38; and TFD6, and TFD20. This was not surprising, since these appeared
identical by all other measures.

Several other strains had nearly identical RFLP pattems, differing by

60

.33 E vmmmsumé <ZDu m3 .0 mourn—Em co v83 58% up: 2 8:88. .86 EUR? .8 6.85» 2.: 2 “£3 8:82 ..

 

 

w and? @5223 mugceouaunmgoam 09925 8.890% Emmi—om 55.—m :2
v 93?. .8 mezoscmfizum mam" Jm .o $6.4 .8me 352m
2. 2&2 .8 388838 8852 588m 35:5 2.8
w «an? .8 mueosemfiim onto—25m «tun—om 35:0 n.2—
v SE2 .8 mueoEamxium Bot—mm topaz «53:02 “0+.N
a 9&2 .8 3858:38 481.329 amaze: nix
«8860 .8 $5.583: 968.32 comp—O w:
H Sum maudzufitx Hucuxouoxm o&o£_=m 3.830 88

N Sum mzfizuetk Hecmxougz $2 :16 .683. 5322833 ROE.

N 58 .8 8835.8 32 :1 a .983 58222 Ne...
N 58 .8 «Sagan 83 is 8&0 88:0 88::—
N 58 .8 3835.8 :2 ..__w a 58$ 58222 89:.
N 58 .8 amassficam 32 :3 a. .988. amaze: we...
N .28 .8 3.838% 82 .315 885 85—
N Sum 338.8 etwfiefitam $3 56:3 use 53$ 39:80
N 5mm 35:258. nucuxoucxm 33 :3 .u .628. 535:2 mNDE.

N Sum 2:323» 8533.5; vam— :16 .980... cumin“: 89:.

N Sum 3:323» muemmmuu; «may in .m .028. £953.; 39E.

N Sum $333838“ «28336? $3 :3 .u .683. 5365.33.88 8%.

N 28 .8 3333.8 32 2.. a .3. 88522 Ex
N 58 .8 sassfism 32 .a a .88... 8328388 $92.
N can .8 «53388 BE .3 .m 55. 886 832....—
N 93 3:883 Riemann? £3 £23588 can :00 wasfisa‘ «was:
mafiwasm "*2 3:56 .38 35.8qu Euro 555m

§a~n§u°¢

 

420. m: we 985%: .8 was .58.. mice»... 9.3 A «:3.

61

100 BP Ladder
100 BP Ladder

    

.—

Figure 1. Amplified 16S ribosomal DNA restriction analysis (ARDRA) of
2,4-D degrading isolates. DNA was PCR amplified with 16S rDNA primers,
digested with four-base cutting restriction enzyme, and size fractionated by
electrophoresis on a 1.5% agarose gel. Figure 1a digested with MspI; 1b with
HaeII; 1c with Sau3A; 1d with AluI; and 1e with Rsal.

62

.ouug mm oow

m0
07mm
mém
rs.
mém
3 out.

 

Figure 1b. Digested with HaeII.

63

.5261. am 00*

0
O
5
1

_

.. “w- ~.,.. “d

- 600

300

100

 

Figure 1c. Digested with Sau3A.

64

52:3 n_m 00—.
.eoo .m

mém

9‘ 52m
FOLEN

070m

94.—

Ow<m

gas...
.88.. am 8F

0
0
5
4|
_

- 600

- 300

- 100

 

Figure 1d. Digested with AluI.

65

528.. am 8,
s8 .m

98

8. 52m
Pozwd

0 Tom

was?

ow<m

gas:
52.3 am 9:

 

Figure 1e. Digested with RsaI.

66

only one or two restriction fragments. For example, IMP134 was closely
related to TFD41 and TFD38; D5 was nearly the same as TFD6; and 86-9
differed from TFD23 and TFD31 by only one AluI band.

The fragment sizes resulting from these five restriction enzyme
digestions were combined and 18 unique strain types could be distinguished
(Table 2). The maximum likelihood method of Felsenstein (1992) using
RestML program was used to estimate the phylogeny from restriction data.
The resulting tree (Figure 2) contains a topology similar with some
differences to that obtained by the sequence data (discussed below).
Diagnostic restriction fragment bands: These restriction fragment length
polymorphisms were examined to see if any were diagnostic for particular
lineages. We found several restriction fragments which may be diagnostic for
different lineages including three for the Proteobacteria, or at least the alpha,
beta and gamma subdivisions examined here; one band diagnostic of the beta
subgroup; and three diagnostic restriction fragments for the alpha subgroup.

The several restriction fragments shared by all strains included a 150 bp
RsaI fragment; a 200 bp AluII fragment; and a 100 bp MspI fragment.
Although it is tempting to speculate that these are signature fragment sizes
for the Proteobacteria, more strains would have to be tested.

Only one band, a 350 bp HaeIII fragment, appeared to be diagnostic for
all of the beta Proteobacteria group 2,4-D degraders in this study, and was not
found in any of the alphas, nor in E. coli, a member of the gamma
Proteobacteria.

One restriction fragment, an 870 bp Sau3A fragment, although found

in all beta strains tested here, was also found in two of the alpha strains,

67

Table 2. Presence, +, or absence, 0, of restriction fragments of amplified 16S rDNA.

 

68

Figure 2. Dendrogram of 2,4-D degrading bacteria based on

K712

B

EML146 cc

8610

865

2401

[(1443

TFD39

TFDZB

TFD31

a;

an

BR16001 B

TFDBBB

TFD41

8

1501346

E. coli

maximum likelihood analysis of restriction data using tetrameric

restriction enzymes. Note that branch lengths are not to scale.

69

thereby lessening its usefulness as a diagnostic band.

Several bands were found in all of the alphas, but not in the betas nor
in E. coli. These fragments included a 350 bp fragment from MspI, 150 and
300 bp fragments with HaeIII, and a 150 bp fragment from Alu I 150.

168 rDNA sequencing: A flowchart for acquiring and analyzing the 16S
rDNA sequence data is depicted in Figure 3. Nucleotide sequences from
approximately 510 to 0 (relative to Escherichia coli) were obtained for all
strains. Since nucleotide resolution began to deteriorate in some strains from
approximately 150 to O in the reverse direction, this data was not used in
subsequent analyses. An example of the Chromatogram obtained from the
ABI sequencer depicts the typical data quality of sequences (Figure 4). The
quality of the data for the first 300 nucleotides was high since the average
numbers of ambiguous nucleotides averaged 1 per 300 nucleotides with a
range of 0 to 3. As a control for sequencing accuracy, strain TFD6 was
sequenced twice on different dates and yielded identical sequences on both
occasions. In addition, strains TFD6 and TFD20, which were identical in
every other respect (Tonso et al., 1994; Fulthorpe et al., 1994) also showed
identical sequences, as did TFD38 and TFD41. In addition, some strains that
had minor difference in REP patterns contained identical 16S sequences. For
example, TFD23 and TFD31 had a few different REP bands, but shared
identical 16S rDNA sequences. Also, E2w1 contained a REP pattern slightly
different from that of TFD6, yet both had identical 16S rDNA sequences.
Sequence alignment: The 165 rDNA sequences were aligned manually using
Squd and MacClade (Figure 5). Additional 165 rDNA reference sequences
from the Ribosomal Database (Larsen, 1993) and from GenBank were added to

70

.53 85:3.“ m3 mo £92me 98 838:8 o5 wan—365 tanuzofi .m Eswfi

88555055580 IIII'
mwmmmnmflmmmw Emuweucmfl 38280
mDm «not mmuzoswmm wfimD 4/
mmUfiHmswmm cwMfi/IV UUOU‘F‘dBBUOBFFrOOUU
m mam umfiflm
bmmumm wfiuamsmmm
\(\/
8H5 ..

/t b J
He + HQH mHmeHnH
£82 mom

71

Q8538 3382:“ <mum 388 mamamxmoa 32%? 69a Eauwofifiofiu no «.mfiaxm .v «2&5

“J: ’43? 3.3“=.ééaé .

_

.Ez. u . «mug—ooéasufi ”.333: seguttzmcwo? 3 2380 2303653023 ~3008§e¢<32§oouogp3nu38u5§383—3“

3:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I

33?;3.”54:.Mfiwumfldfimwaffisfliimwéwsuflflzjmga‘éga.1%.?3$3.40:. .

:Kmmm 3.. 238029.53. _ 03§<Sw$<¥. . GSTSQOWS. popt<§8§2~§~8~ofi 889:5: .38? pgaxfiuroii

 

 

 

 

 

 

 

‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l
[

Burkholderia
Burkholderia
Burkholderia
Burkholderia
Burkholderia
TFDG

BRIGOOI
EML159

TFDZ
Burkholderia
Burkholderia
RASC

K712

JMP134

TF041

TFDBB

cepacia
cepacia DS
cepacia G4
Hallei
gladioli

caryophylli
andropogonis

Alcaligenes eutrophus

Burkholderia solanacearum

TFD39

Alcaliganes xylosaxidans

TFD9
869
TFD23
TFD31

Rhodofianmx fermentans
Cbmamanas testostosteroni
Halamanas elongate

E. 0011

Halamanas halmph

Dal.hphila
'I-18‘
2401

Sphingomanas adhaesiva

Sphingamanas,paucimobilus
Flavobacterium capsulacum

865
3610
K1443
Eflld46

Sphingnmonas yanoikuyae

Rhodqpseudamonas‘palustris

M1

Desulfanonile tiedjii

1

72

2

3 4 5

6]
123456789012345678901234567890123456789012345678901234567890]

TCGGGCCICGCGCTATAGGGPIGGCCGATCGCIGATPAGCPAGPIGGTGGGGPAAAGGCC

....? .......................................................
.................................................. C.........
AG..A ...... ?...C....GC .......... A ...........................
CG..A .......... C....GC .......... A ...........................
AG..A ...... ?...C....GC .......... A ...........................
..A ......... CA. .AC .......... A ...........................
.............. CA....GC..........G...........................
A ............... GC .......... A ...........................
A .A ............... GC .......... G ................ A ......... T
....A ...... ? ........ ?C ........ ? .................. A ......... T
AAC.T ........ G ..... AGC ........ T ........................ GA
CAA.C ...... ?.AGC...AGC A ...... T .............................
CAA.C ...... ?.AGC...AGC.A ...... T .............................
?AA ............ C...AGC ........ T .............................
CAA ...... AT ........ AGC ........ T .............................
....?...T. ?. .CATA A ..... ?.T ........................ T .
CAAmA...T..A....T..AGC ....... AT.G ...................... C...T
CAA.A...T. A .TA.AGC ....... AT.G .................... GGC ..T
CAA.A ........ G..T..AGC ....... AT A .................... GG.A..
CAA A ........ G T..AGC ....... AT A .................... G.mA..
CAA A ........ G. T .AGC ....... AT‘A .................... G ‘A..
CAAMA ........ G T..AGC ....... AT.A ................ A ......... T
........ T.T.. CTA.AGC T......Am.....G...................T
A .T ....... C..A.GA TT .T.G ........ G ....... A ..... C ..T
......... 'I'I' CCAGTCGA'IGGAC ..T
..A...T ....... C..A.GA..TT.. T G ................ A. ..... C ..T
..A...T ....... C..A.GA...T.. T.G ................ A ..... T. .T
G..A...T ....... C..A.GA...A..?T.G ............... A ..... GC ..
AAA.ATA.AT T.GCAG. A.GA...CGC.TAG ...... G ............... T T
AAA.ATT AT ..CCAG. A.GA...CGC.T.G ...... G ....................
AAA.ATT.AT ..CTG.. A.GA...CGC.TTG ...... G ....................
AAA.ATT.AT .GC.A.. A.GA...CGC.TAG ...... G ....................
AAA.ATA AT T.GCA.. A.GA...CGC.T%G ...........................
AAA.ATA.AT.T.CCAG..AMGA...CGC.TAG ................ A ......... T
AAA.ATA AT T.CCAG. A.GA...CGC.TAG ................ A ......... T
AAAWATT'AT T.CCAG. AWGA...CGC.TAG ................ A ......... T
AAAWATT'AT ..CCA...A.GA.. CGC.TAG ................ A ......... T
AAAWATT'AT ..CGA.A.A C .CGC.T .................. A ..... T ..T
AAAmAflTEAT T.GGA.A,A.C ..CGC.T .................. A.T C ..T
-GAA ..GATGC A -. .A.G TCGC .ACC ....... T ....... A ..... C ..T

[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
[60]
(60]
[60]
[60]
[60]
[60]
[58]

Figure 5. Alignment of 165 sequences from positions 209-506 according to the E. coli
numbering system.

73

I 1 ]
I 6 7 8 9 0 1 2]
[ 123456789012345678901234567890123456789012345678901234567890]
Burkhol deria cepacia TACCAAGGCGACGATCAGTAGCT?GTCIGAGAGGACGACCAGCCACACIGGGACIGAGAC
Burkholdéria cepacia DS ....... ? ............... G ....................................
Burkholderia cepacia G4 ....................... G ....................................
Burkboldbria mallei ....................... G ....................................
Burkholdéria gladioli C ...................... G ....................................
TFD6 ....... C ........ T. . .T. .G ....................................
BRIGOOl ....... C ........ T...T..G ....................................
841.159 ....... C ........ T. . .T. .G ....................................
TFDZ ....... C ........ T. . .T. .G ....................................
Burkholdéria caryophylli ................ C ...... G ....................................
Burkholderia andrqpogonis ................ C ...... G ....................................
RASC C ............... C ...... G ....................................
K712 C ...... ? ........ C ...... G ....................................
JMP134 C ...................... G .............. T .....................
TFD41 C ...... C ............ T. .G .............. T .....................
TFD38 C ...... C ............ T. .G .............. T .....................
Alcaligenes eutrophus ....................... G .............. T .....................
Burkholderia solanacearum C ...................... G .............. T .....................
W39 C ...... C ........ C ..A..G .............. T .....................
.Alcaligenes xyiosoxidans C ............... C ...... G..T .................................
TFD9 C ...... C ........ C ...... G. .T .................................

869 ....... C ........ T. . .T. .G ........................... A ........

'I‘FDZ3 ....... C ........ T. . .T. .G ........................... A ........

TFD31 ....... C ........ T. . .T. .G ........................... A ........

.Rhodoferax fermentans C ............... T ...... G ........................... A ........

Cbmamonas testostosteroni ....... C.TG ..... T ...... G ....................................

Halamonas elongate C ............... C ...... G ........... T..T ........ TC ...........

.3. coli C ..T ........... CC ..... G ........... T ............... A ........

Halamanas halmph ................ C ...... ? ........... T..T ........ TC ...........

Del.hphila c ........ CG ..... c ...... G ........... T. .T ........ 'nc ...........

'I-18' ....... C..? ..... C...C..G ........... T..T ........ TC .......... ?

2401 ...... CC ........ CT ..... G ........... T..T .....................

Sphingamanas adhaesiva ....... C ........ C ...... G ........... T..T .....................

Sphingomonas‘paucimobilus ....... C ........ CA ..... G ........... T..T .....................

Pfiavobaccerium capsulatum ....... C ........ CT ..... G ........... T..T .....................

865 ................ CT ..... G ........... T..T .....................

8610 C ...... C ........ CT..T..G ........... T..T .....................

X1443 C ...... C ........ CI‘ ..... G ........... T. .T .....................

M1146 C ...... C ........ CT ..... G ........... T. .T .....................

sphingomonas yanoikuyae C ............... CT ..... G ........... T..T .....................

Rhodqpseudamonas‘palusCIis C ...................... G ........... T..T ........ T ............

M1 C ...... C ............... G ........... T .T ........ T ............

DasuIIOmonile tiedjii C..? ..... CGA...GGT ..... ? ........... .T ............ A ........

Figure 5. (continued)

[120]
[120]
[120]
[120]
[120]
(120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[120]
[118]

[
[
[

Burkholderia
Burkbolderia
Burkholderia
Burkholdbria
Burkbolderia
TFD6

BRIGOOI
EML159

TFDQ
Burkholderia
Burkholderia
RASC

K712

JMP134

TFD41

TFD38

cepacia
cepacia D5
cepacia G4
mallei
gladioli

Alcaliganes eutrophus

Burkholderia solanacearum

TFDB9

Alcaliganes xyiosaxidans

TPDS
B69
TFDZB
TFDBI

Rhodoferax femmentans

1
2

74

3 4 5

6

l
7 8]

123456789012345678901234567890123456789012345678901234567890l

ACGGCCC?GACTCCTACGGGAGGCAGCAGTGGGGAASTTTGGACAATGGGCGAAAGCCTG

....... A....................................................
....... A....................................................
....... A............................................C.......
....... A....................................................
....... A............................................C.......
....... A............................................C.......
....... A............................................C.......
....... A..........................................G.C C
....... A..........................................G.C C
....... A............................................C.......
....... A....................................................
....... A............................................C..C..
....... A..........................................G.C..C..
....... A..........................................G.C..C..
....... A..........................................G.C..C..
....... A..........................................G.C..C..
....... A..........................................G.C..C..
....... A.............................A......................
....... A......?...................................G.. C
....... A..........................................G.. C

Cbmamanas testostosteroni ....... A ....................................................
Halamanas elongata ....... GA ............................ A ............ G.C..C
E. coli ..T..A ............................. A C .......... C .......
HHJamanas halmph ...... ?GA ............................ A ......................
Dal.hph11a ..... ???A....? ......... ?.? ........... A ......................
'1-18‘ ...?...GA ..... ? ...................... A ............ G C C
2401 ....... A ............................. A ......................
sphingumnnas adhaesiva ....... A ............................. A ......................
Sphingamonas‘paucimobilus ....... A ............................. A ......................
FTavubacterium capsulatum ....... A ............................. A .............. C .......
865 ....... A ...................... A ...... A. ............ GTC..C
B610 ....... A. C .CC. C ............... T..A ................. A
K1443 ....... A ...................... A ...... A ............ GHC..C
EML146 ....... A ...................... A ...... A ............ G C C
Sphingamanas yanoikuyae ....... A ............................. A ......................
Rhodqpseudamanas‘palustris ....... AA ............................ A ............ G C
M1 ....... AA ............................ A .............. C .......
Dasulfhmonile Ciedjii ....T.??.? ..................... A ..... A CG ...... AG....CT..

Figure 5. (continued)

[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[180]
[178]

75

[ 1 2 1

[ 8 9 0 1 2 3 4]

[ 123456789012345678901234567890123456789012345678901234567890]

Burkholderia cepacia AICCAGCAATGCCGCGTGTGIGAAGAAGGCCI‘ICGGGI'IGTAAAGCACI‘ITIGTCCGGGA [240]
Burkbolderia cepacia D5 .......... C ............................................... A. [240]
Burkholderia cepacia G4 .......................................................... A. [240]
Burkholderia mallei .......................................................... A. [240]
Burkholderia gladioli .......................................................... A. [240]
TFD6 .......... C ............................................... A. [240]
BR16001 ..... ?. . . .C .............................. ? ................ A. [240]
1141.159 .......... C ............................................... A. [240]
TFUZ .......... C ............................................... A. [240]
Burkholderia caryophylli .................... C ..................................... A. [240]
Burkholderia andropogonis ...................................................... T ..... [240]
RASC .......... ? ............................................... A. [240]
K712 .......... C ............................................... A. [240]
JMP134 .......................................................... A. [240]
TFD41 .......... C ............................................... A. [240]
TFDBB .......... C ............................................... A. [240]
Alcaligenes eutrophus .......................................................... A. [240]
Burkholderia solanaceannn .......................................................... A. [240]
TFD39 ....... C. .C ................................................. [240]
Alcaligenes xylosoxidans ....... C. .C ......... C. .T ..... T.G. - ................... G.A. .A. [239]
TFDS ....... C..C ......... C..T ............................. G.A..A. [240]
369 .................. CAG. .T .................... C‘IG ....... A. . .A. [240]
TPDZ3 .................. CAG..T .................... C'IG ....... A. ..A. [240]
TFDBl .................. CAG. .T .................... C'IG ....... A. . .A. [240]
Rhodoferax fermentans .................. CAG. .T .................... GIG ....... A. . .AG [240]
Canamanas testostosteroni .................. CAG. .T ....... C ............ C'IG ....... A. . .A. [240]
Halanonas elongate: ....... C ....................... C ................... CAG.GA. . . [240]
E. coli ..G....C ........... A ......................... T ..... CAG.....G [240]
Halamnas halnph ....... C ...... ? ................ C ................... CAGTGA. . . [240]
Del.hphi1a ....... C ........................................... CAG.GA. . . [2401
'I-18' ..... ?.C..C ........................................ CAG....GA [240]
2401 .......... ? ....... A....T ......... A ............ T ..... AC.....A [240]

Sphingomonas adhaesiva .................. A. . . .T ......... A ............ T ..... AC ...... (240]

Sphinganonas paucimobilus .................. A. . . .T ....... C.A ............ T ..... AC ...... [240]

Flamcterium capsulamm .................. A. . . .T .............. C ....... T ..... AC.A. . . ? [240]

865 .................. A....T ....... C.A ............ T ..... AC...A.. [240]

8610 .......... C..C....A....T ......... A ............ T ..... AC ..... G [240]

K1443 .................. A. ...T ......... A ............ T ..... AC ...... [240]

311.146 .................. A. . . .T ......... A ............ T ..... AC ...... [240]

Sphinganonas yanoikuyae .................. A....T ......... A ............ T ..... AC ...... [240]

Rhodopseudanoms palustris ....... C .......... A. . . .T ....... C.A ............ T ....... G ..... [240]

Ml ....... C .......... A....T ....... C.A ............ T ....... G ..... [240]

Desulfanonile tiedjii .CG ...... C ........ G ........... T ............... C. .G.CAGG ..... [238]

Figure 5. (continued)

[
[
[

Burkholderia
Burkholderia
Burkholdbria
Burkholderia
Burkholderia
TFD6

BR16001
EML159

TFDZ
Burkholderia
Burkholderia
RASC
K712

JMP134

TFD41

TFD38
Alcaligenes eutrophus
Burkbolderia solanacearum
TFDBS

Alcaligenes xylosaxidans
TTDB

369

TP023

TFD31

Rhodoferax fezmentans
Cbmamonas testostosterani
Kalamanas elongata

E. coli

salamanas halmph
Del.hphila

'I-18'

24D1

Sphingamanas adhaesiva
Sphingomonas‘paucimobilus
Ffiavobacterium capsulatum
865

8610

K1443

EML146

Sphingumonas‘yanoikuyae

capacia
cepacia DS
cepacia G4
nallei
gladioli

canyophylli
andrqpogonis

2
4

76

5 6

7 8 9

]
]

12345678901234567890123456789012345678901234567890123456789]

AG-AAATCCI'IGGCI‘CTAATACAGCCGGGGGATGACGGTACCISGAAGAATAAGCACCGG

 

 

 

 

Rhodqpseudamonas[palustris .

M1
Desulthmonile tiedjii

Figure 5. (continued)

 

 

 

 

 

 

- ..................... T ..................................

- ..... A. CTGG ....... CCGGA.T ..............................

- ....... GA.GG ...... TCCTTC ................................

- ..A.. C.TGGT ...... CCGT .................................

- ..A.. C.TGGT ...... CCGT .................................

- ..A.. C.TGGT ...... CCGT .................................

- ..A...CC.TC ....... G TG .................................

- ....... GCCTGA ...... C GGC ................................

- AGGCGA.GG ...... TCIHTT‘CT .............................

- ...... C. .TC ...... TG? ...................................

- ...... C...TC ...... TGC ...................................

- ..A..G.T..C ....... G.TG .................................

- ..A...CC..C.G ..... G.TG .................................

- ..A...CC..C.G ..... G '1‘] .................................

- GGC.CTGGT ...... CCGG..TC .............................

- ..... GCACTGGT ...... CTGGT T ........................ G .....

- ..G..A ..C ........ TTG.T ................ C ..............

- CGECGT.GGT ...... CC.GC CAC .......... T C ..............

..- CGTCAT.GGT ...... CC.GT AAAC .......... T C ..............

C.- CGG..CT ........... GG .CTA ............ T ...............

C.-. CGG..CT ........... GG..CTA ............ T ...............

C.-. CGG..CT ........... GG..CTA ............ T ...............

C.- ..GG...CT.- ....... GAGG..C.'I'C ............ T ...............

C.- ..AG.C GG ...... TCC ..... TC ............ T ...............

..- 'lG T GTCGGT ...... CCCBCAA..GA ATC .‘IC.C.. G .........

T.- GGGAG.AAAGT ...... CHET'CTCAT ..... T....C.C.. G .........
- GG T.GTCGGT ...... CCGGCA..A G .ATC..TCAC.. G C

- CG T GT GGT ...... CCTGCAA.A G .ATC..TC.C.. G .........

..- GG GC'IG" C ..... CCAGCA..A G ..A‘IC..'IC.C.. G .........
T ATA ..... A ..... T G ......... T
T ATA ..... A ....... G ......... T
. ATA ..... T ....... G ......... C
T - ATA ..... A ..... T.G ......... T
T -- ATA ..... A_ T ..G ......... C
. -------------------------- ATA ..... T ....... G ......... T
T ATA ..... A ....... G ......... C
T ATA ..... A ....... G ......... C
T ATA ..... A ....... G ......... T
ATA ............ C .......... C
. .ATA ............ C .......... C

..A..CCTTZGC.GA.G....GTC GT.AAAC ........... CC A 66 .........

[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[298]
[297]
[298]
[298]
[298]
[298]
[297]
[298]
[298]
[298]
[298]
[298]
[298]
[273]
[273)
[273]
[273]
[273]
[273]
[273]
[273]
[273]
[273]
[273]
[297]

77
the alignment. For the alignment of 16S rDNA sequences, only the
nucleotides from 506R to 209 (using the E. coli numbering system) were used,
and a mask was used to exclude any ambiguous nucleotides. This resulted in
283 aligned nucleotides which included 95 variable nucleotide positions.
Phylogenetic inference: Two methods were used to infer the phylogeny of the
strains. Once the sequences were aligned and ambiguous sites eliminated,
percent differences in nucleotides between strains were used to make a
pairwise matrix of distances (Table 4) between strains sequenced for this study
and sequences from Genbank (Table 3) used as references. This matrix was
then used in the Neighbor Joining distance matrix program of Felsenstein to
create the tree depicted in Figure 6.

In addition to using a distance matrix method for creating a tree, a
parsimony method was also employed using PAUP (Swofford, 1993) (figure
7). In this analysis, the beta group contained 25 taxa with 68 phylogenetically
informative characters. The alpha subgroup contained 10 taxa which had 34
phylogenetically informative characters. Within that, the eight
Sphingomonas-like strains had 17 phylogenetically informative characters.
Confidence level for tree: The validity of using partial 16S sequences for
making reliable trees has been established by Lane et al. (1985) who found that
fewer than 350 nucleotides of aligned sequence data yielded the same tree as
did using the entire 165 sequence, for which only about 950 nucleotides were
aligned unambiguously (Lane et al., 1985). In this study, strains which
differed by more than 5% could be unambiguously grouped. It was also
found that some of the uncertainty of the phylogeny of selected beta

Proteobacteria was the same using either 300 or 1400 nucleotides (data not

78

 

 

.282 22883.8 .2

2. 3.233.348 .2
2. on 3.2.07 Causal: .an
2. 2 2 2-2 .2
2. 2 2 2 28.- .2
2. 2. 2. 2. 2. 2.32:3 381823282 .2
2. 2. 2. 2. 2. 2 2 .2
2. 2. 2. 2. 2. 2 2 229231 4:88:22. .2
2. 2. 2. 2. 2. 2 2 2 23..- .2
2. 2. 2. 2. 2. 2 2 2 2 2.2 .2
2. 2. 2. 2. 2. 2 2 2 2 2 .2 .2
2. 2. 2. 2. 2. 2 2 2 2 2 2 322288— .33829»... .2
2. 2. 2. 2. 2. 2 2 2 2 2 2 2 22 .2
2. 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2.223.... 433032....» .2
2 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2 32:82... 223382.. .2
2 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2 2 22 .2
2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 33222 x2283: .2
2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 22.» .2
2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 282-32» .3318 .2
2. 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 22. .2
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 263.82.? 229282 .2
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 528-33.. 2.3.32»..- .2
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2.4498:- 223103 .2
2. 2. 2. 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2 22-. .2
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2 2 238 .2
2. 2 2 2 2 2. 2. 2 2. 2. 2. 2. 2. 2. 2. 2. 2 3 2 2 2 2 2 2 2 20:33 2.3.22.3- .2
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2 2. 2. 2. 2 2 2 2 2 2 2 2 2 2 238398..- c‘uinofiuan .2

2. 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2 2 2 2.3.8.2.... 22335:.- .2
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2 2. 2. 2 2 2 2 2 2 2 2 2 2 2 2 22..- .28322. .a
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2 2 2 2 2 3 .288 422.32.3- ..
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3.22.2... 2.5.32.5. .2.
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 3 3 2 2 2 2 2 2 2 2 3.. 2 2.32... 422.3225 .-
2. 2 2 .2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 3 2 2 2 2 2 2 2 2 2 2 2 22v .2
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 22 .c
2. 2 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 82 .n
2. 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 ... 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 22 .u
2. 2 2 2 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2. 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 SI. .2
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 a o 2. u a c a a u o ...-.8

£35.52 ounce—us: 35.? man .8 ‘22— bm—uafifi 8:252 .522— 3252: 5.22: 362.5 d 63:.

79

Table 3. Reference strains used in building phylogenies.

 

GenBank Accession

 

Strain Numbers
Burkholderia cepacia G4 1.28675
Burkholderia gladioli X67038
Burkholderia solanacearum X67041
Burkholderia mallei 555000
Burkholderia cepacia M22518
Burkholderia andropogonis X67037
Alcaligenes xylosoxidans M22509
Alcaligenes eutrophus M32021
Rhodoferax fermentans D16211
Burkholderia caryophylli X67039
Sphingomonas sp. D16148
Sphingomonas capsulata D16147
Sphingomonas yanoikuyae 013782
Sphingomonas paucimobilus D16144
Sphingomonas adhasiva D16146
Sphingomonas parapaucimobilus X72723
Plavobacterium capsulatum M59296
Blastobacter Z23156
Rhodopseudomonas palustris D25312
Desulfomonile tiedjii RDP"
Halomonas elongata RDP"
E. coli RDP"

 

" 16S sequences taken from Ribosomal Database Project

 

TFD6
EML159
BR16001
TFDZ

m Burkholderia caryophylli

RASC

K712

Burkholderia cepacin
Burkholdcrin cepacia D5
Burkholderia mallei
Burklzolderia gladioli

B urkholderia andropogonis
Comamonns testostosteroni
B69

TFD23

TFD31
Rhodoferaxfermentans

TFD38

Alcaligenes eutrophus
Burklwlderia solanaccnrum
TFD39

Alcnligenes xylosoxidans
TFD9

Halomonas elongatn

Del. lmlophiln

24D1

F luvobacterium capsulatum
Sphingomonas adlmesiva
3610

Sphingomorms ymzoikuyae
B65

 

K1443
EML146
",' ' u paucimobilus
,, M1
8 Rhodospeudomwms palustris

 

...
i—l

Desulfomonile ticdjii

Figure 6. Dendrogram of 2,4—D degrading strains (bold type) and reference strains

(italics) based on Neighbor Joining Method of PHYLIP. Subgroups of Proteolmcteria are

indicated in color: blue indicates betas; green gammas; red are alphas; and black

indicates delta Proteobaclcria. Numbers along branches indicate unambiguous nucleotide

base changes between branches.

81

Burklmldcria ccpacia D5
Burkholdcriu cepncia C4

 
    
  

TFD39

 
 
   
   

Burkholdcrin maIlci

llurklmldcria gladinli
BRI600]

TFD6
EM L159

Burk/ioldvrin cnryophylli
Burklmldurin andropogonis

RASC

K712

B JMP134
TFD38

_ c Burklmldcria sulmmccnrum
Alcaligcncs eutmplms

Alcaligcncs xylnsoxidnns

 

  

 

ledqfcraxfermen tans
" testos tos terom'
Halomonas clongata

Y Del . halophila
I-18

 

Ecoli

24D1

Splzingomonas adhaesiva
Sphingomonas paucimobilus
Flawbacterium capsulatum

  
 
 
 
    

B65

  
 

8610
K1443

EML146
Sphingummms ymzoikuym'

RliodnpsczIdonionas paluslris
M1

   
 
 
  

 

 

Desulfomonilc tiedjii

 

Figure 7. Phylogeny, based on maximum parsimony, of 16S rDNA from 2,4—D
degrading bacteria, indicated in bold letters, and reference strains, indicated in
italics. Colored lines indicate phylogenetic groups of Proteobacterim blue are beta,
green are gamma, red are alpha, and black is delta Proteobacteria.

82

shown).

There has been discussion as to whether or not 300 nucleotides is
sufficient data to adequately resolve a phylogeny for a group of molecules.
According to Stewart (1993), as a general principle, phylogenetic analysis
works well only when there are more phylogenetically informative characters
than taxa (strains) to test. This was certainly the case in this study, since the 23
strains under investigation had 95 variable characters.

The 300 nucleotides of aligned partial sequences used here yielded a
tree with branching patterns similar to several trees published previously.
For example, the tree produced by maximum parsimony correlated well with
the same major branches as the recently released tree in the RDP, version 4,
and with the previously published tree of Woese, 1984, which used oligo-
nucleotide cataloging to build a tree which included members of the alpha,
beta and gamma Proteobacteria (or group R1, R11 and R111 as they were then

designated). The most parsimonious tree presented here also agrees well
with that of Yabuuchi et al., depicting a group of beta 2 Proteobacteria, which
is now referred to as the genus Burkholderia (Yabuuchi et al., 1992; 1188,
1992). The tree presented here also has a similar topology to that of the beta
subgroup published by DeWhirst et al., 1989. Our tree also had a similar
branching order as that for the the Sphingomonas and Rhizomonas group of
alpha 4 (rRNA Superfamily IV) Proteobacteria (270 aligned nucleotides,
positions 50-350) determined by Van Bruggen et al., 1993.

83

DISCUSSION

This study shows that the 2,4-D degraders investigated here represent
diverse members of the Proteobacteria (Stackebrandt et al., 1988) which
encompasses a large proportion of all Gram-negative bacteria (Murray et al.,
1990). Proteobacteria aptly describes this diverse group of strains, since the
term "proteo" is from the Greek sea god, Proteus, who had the ability to
assume many diverse shapes and the Proteobacteria have members with
many diverse metabolic capabilities.

Two methods based on comparisons of 16S ribosomal genes were
employed to examine the diversity of these strains. The RFLP method was
found to be useful for predicting minor differences in closely related strains
and the 16S sequencing method was found useful in placing these strains
among the burgeoning branches of the 16S phylogenetic tree (W oese, 1994;
Larsen, 1993).

RFLP analysis of amplified 16S rDNA

This study compared 16S rDNA RFLPs to 168 partial sequence data.
The trees derived from the two sets of data correlated well at the "twigs" of
the trees, but not at deeper branches. The diversity of strains observed by
sequence data generally reflected the diversity of strains determined by
ARDRA data. For example, strains TFD23, TFD31 and B6-9 shared identical
sequence similarity over all 300 nucleotides. However, 86-9 lacked one
restriction band (a 350 bp Rsal fragment) that was shared by TFD23 and TFD31.
This difference is probably due to base pair differences in sequence outside the
300 nucleotides that were sequenced.

84

For initial quick and inexpensive genomic screening of many strains,
where sequencing is too expensive, ARDRA has been shown to predict
similar 168 genes and is therefore useful in eliminating duplicate strains
before sequencing. ARDRA is better technique than REP-PCR for this
purpose, because ARDRA data is reproducible under all conditions, whereas
REP PCR can give variable results depending on the concentration of DNA,
of magnesium, or of primer and on a variety of other conditions (Ausubel et
al., 1992). ARDRA has the additional benefit over REP-PCR in that a database
of ARDRA fragments can be made with which to compare test strains.
Because REP fragments can be elements involved in recombination (Higgins
et al., 1988; Stem et al., 1984; Gilson et al., 1987), at least some fragments may
be subject to rearrangement within a few generations of laboratory cultivation
(Nakatsu, personal communication) making REP-PCR suitable for strain
identification at a more refined level than that provided by 16S rDNA RFLP
analysis. This was observed in strains TFD23 and TFD31 which have identical
16$ rDNA sequences, but showed different REP patterns (Tonso, 1994;
Fulthorpe, personal communication).

Although ARDRA provides an inexpensive initial screen for choosing
which strains to characterize more fully by sequencing, the information it
yields is not suitable for phylogenetic analysis for five reasons: 1) the data
cannot be compared to reference data outside this study because there is no
large database of RFLP data as there is for 165 sequence data; 2) more than
five restriction enzymes are required for proper resolution of deeper
phylogenetic branches and eight restriction enzymes may not be enough

(T iedje and Moyer, personal communication); 3) data is prone to error due to

85

band doublets; and 4) ARDRA has fewer phylogenetically informative
characters and 5) bands of similar size which are not necessarily homologous
could possibility be mistaken for convergent bands.

When used in conjunction with ARDRA to eliminate related strains,
analysis of 16S rDNA sequences gives a reproducible, reliable and increasingly
inexpensive means of determining the phylogenetic placement of strains.
Advantages of 16S rDN A sequencing over FAME and BIOLOG, two of the
most popular and widely used methods of identifying bacteria, include 1)
largest strain database (2600 strains in the Ribosomal Database Project versus
1100 for BIOLOG and 1300 for MIDI in 1994); 2) more reproducibility
independent of the strain's metabolic condition; 3) better software programs
for phylogenetic comparison; and 4) characters of 168 rDNA are independent,
unlike phenotypic characters of FAME or Biolog in which a single genetic
change could influence a whole range of genetic characters. These benefits
may justify the higher cost of sequencing especially since the cost of
sequencing is declining.

In this study, we found that the most appropriate use of ARDRA is for
determining the relatedness of strains to reduce replicates before sequencing.
For this purpose, no more than two restriction enzymes would be necessary,
making this technique a cost and time effective initial screen.

Diversity of 2,4—D degraders

Thirteen unique species of bacteria were observed in this study. These
unique strains all differed by more than 3% from each other and from the
strains in GenBank. Sequences diverging by more than 2.5% at the level of
16S rDNA sequences have been found to correspond to total DNAzDNA

86

sequence differences greater than 30% (Stackebrandt and Goebel, 1994), which
is the maximum amount of DNA divergence for two bacteria to still be
considered members of the same species (Wayne et al., 1987).

The 2,4—D degraders in this study clustered into six relatively deeply
branched groups which included eight strains related to the genus
Burkholderia, three related to Alcaligenes eutrophus, three related to
Rhodoferax fermentans (which is close to the class Comamonadaceae) one
related to A. xylosoxidans five related to the genus Sphingomonas, and one
related to Rhodopseudomonas palustris.

The majority of the bacteria surveyed here were most closely related to
those of the genus Burkholderia (Yabuuchi, et al.1990) in the beta 11 subgroup
of Proteobacteria. The genus Burkholderia was named after W. H.
Burkholder. Burkholder discovered that onion rot was caused by
Burkholderia cepacia (formerly Pseudomonas cepacia), the genus type strain
(Yabuuchi et al., 1992). All strains in this group, which include TFD2, TFD39,
TFD20, TFD6, BRI6001, EML159, RASC and K712 shared at least 92% sequence
similarity with each other.

Within the Burkholderia group there were several tightly clustered
groups including RASC and K712, which differed by only 1.7%, (five of 300
nucleotides). Strains TFD6, TFDZO, and EML159 were identical. Strain
BRI6001, found in Canadian peat, differed from these strains by one
nucleotide. One strain, TFD39, was most closely related to this Burkholderia
group, and Burkholderia cepacia in particular (93% similarity) although
parsimony analysis showed it had diverged more than other strains within

this group, making it the most difficult strain to place phylogenetically.

87

A group of 2,4-D degraders closely related to the genus Burkholderia
included strains similar to Alcaligenes eutrophus, which differed by seven
percent from strain IMP134 and by seven percent from strains TFD41 and
TFD38. We found that strain TFD41 was most closely related to A. eutrophus
as determined by 16S sequencing, in contrast to its previous designation to the
genus Comamonas, using Biolog and FAME (Tonso et al., 1994). Perhaps this
was due, in the case of Biolog, to the under-utilization of substrates, since
Alcaligenes eutrophus strains typically use 47 of 95 substrates whereas
Comamonas testosteroni typically only uses 25 substrates. Another possibility
to account for these conflicting results is that through gene exchange, TFD41
contains DNA from both A. eutrophus and Comamonas testosteroni-like
parents.

Another cluster of strains found within the beta subgroup of
Proteobacteria included three strains that were most closely related to
Rhodoferax fermentans. R. fermentans, previously known as Rhodocyclus
gelatinosa (Hiraishi et al., 1991) is closely related to the family
Comamonadaceae, members of which have been shown to degrade 3-
chlorobenzoate (Nakatsu, 1992). Three of the 2,4-D degrading strains, TFD31
and TFD23, both isolated from the Kellogg Biological Station, and 86-9,
isolated in Quebec, shared identical sequence and RFLP patterns, except that
36-9 165 rDNA differed in RFLP patterns when digested with RsaI.

Strain TFD9 stood apart among these beta Proteobacteria, and was most
closely related to Alcaligenes xylosoxidans.

One strain, I-18, was found to be closely related to the genus

Halomonas, in the gamma subgroup of Proteobacteria. This strains has been

88

characterized by Dr. Olga Maltseva and found to be a salt tolerant 2,4-D
degrading strain.

The remaining six strains all clustered in the alpha branch of
Proteobacteria. It is interesting that none of these strains hybridized to the
2,4-D degradative gene probes, tfdA and tde from IMP134, as did all of the
beta Proteobacteria (see chapters 3 and 4 for more discussion). This suggests a
deep, natural division in degradative genes for these two branches of bacteria.

One lone strain, M1, which is the most oligotrophic and slow growing
of all the strains used in this study, is 97% similar to Rhodopseudomonas
palustris, a member of the alpha 2 group of Proteobacteria (Wong et al., 1994).
Although M1. does not contain purple photo pigments as does the facultative,
anoxygenic, phototrophic R. palustris, M1 may share the degradative abilities
of R. pal ustris, which is known to grow on a diverse range of aromatic and
heterocych compounds (Sasikala et al., 1994). The slow growing character of
strain M1 correlates well with its placement near the slow growing
Bradyrhizobium, which is 98.3% similar to R. palustris (Wong et al.1994),

The remaining strains all clustered together and were most closely
related to various members of the genus Sphingomonas in the alpha 4
subgroup of Proteobacteria. All but one of the alpha strains were found to be
closely related to the genus of Sphingomonas, named after the lipid
"sphingosine" and sphingoglycolipids which these strains possess. The genus
Sphingomonas was previously classified as Group IV Pseudomonads
according to the classification of Paleroni, et al., (1973). Several members of
the genus Sphingomonas were previously called Pseudomonas (P.

adhaesiva and P. paucimobilus) and Plavobacterium (i.e. P. capsulatum.) All

89

All the Sphingomonas-like 2,4-D degrading strains had colonies with varying
shades of the color of egg yolks, which is the colony color of Flavobacteria ,
whereas the beta, and gamma Proteobacteria observed in this study have off
white to beige colonies.

Relationship to other xenobiotic degrading strains

Comparing the 168 rDNA sequences of these strains to the 16S rDNA
sequences of other known xenobiotic degrading strains may shed some light
on the degradative pathways of these organisms. Several of the strains
sequenced in this study were found to be similar to a species of
Sphingomonas isolated from the the Elbe River downstream from Hamburg,
Germany. This strain of Sphingomonas has been show to grow on dibenzo-p-
dioxin as the sole carbon and energy source (Moore et al., 1993) and
polyethylene glycol (Takeuchi et al., 1993.) Perhaps the genes involved in
dioxin degradation, especially those of the lower pathway, are related to those
for 2,4-D degradation in these strains. Though the Sphingomonas-like
strains in this study have been shown to use the ortho pathway for intradiol
cleavage of catechol like IMP134 (Olga Maltseva, personal communication),
these strains contain 2,4-D genes which are quite different than those of the
well characterized IMP134 (Perkins, 1990) and strains TFD6 and RASC
(Matheson, 1994; Suwa, et al, 1994).

Perhaps these 2,4-D degrading strains contain central metabolic
pathways for degrading aromatic compounds like toluene, 2,4,5-
trichlorophenoxyacetic acid and other aromatic compounds, since similar
strains contain these pathways. For example, several of the 2,4-D degrading

strains in this study were shown to be greater than 95% similar to

90

Pseudomonas cepacia (a.k.a. Burkholderia cepacia) strain G4, an isolate
capable of growth on toluene as sole carbon and energy source (Zhou, Weller
and Tiedje, in Genbank). Another strain, identified as P. cepacia, was found
that could detoxify 2,4,5-trichlorophenoxyacetic acid (Kilbane et al., 1983).

It is interesting that so many of the 2,4-D degrading strains investigated
in this study were related to plant pathogens. Perhaps 2,4-D degradation
phenotype evolved in bacteria known to be plant pathogens. For example,
many 2,4-D degrading bacteria are related to members of the beta group of
Proteobacteria which are all close relatives of plant pathogens, including B.
solanacearum (from potato), B. cepacia (from onion), and P. maltophilia
(barley). Since the plant hosts of these strains produce various aromatic
compounds, including auxin, a functional analog of 2,4-D in plants, it is
possible that the 2,4-D phenotype evolved in organisms associated with
plants.

A group of recently isolated 2,4-D degraders appears to be related to
Xanthomonas (Dunbar, personal communication), a plant pathogen which
contains the yellow pigment xanthomonadin, a brominated poly-aryl ester.
Perhaps the genes involved in the biosynthesis or biodegradation of this
halogenated aromatic compound are related to genes responsible for the
degradation of 2,4-D.

The phylogenetic diversity of 2,4-D degrading organisms is probably
much greater than that indicated here. There are reports of 2,4-D degradation
by strains which are not Gram negative, including such strains as
Arthrobacter (Loos, 1969) and Streptomyces (Speicher, 1954) and other
unidentified gram positive strains (Amy et al., 1987; Holben, personal

91

communication); and even fungi such as Phanaerochaete (Yadav and Reddy,
1994)
Geographic distribution

This study showed that some strains with identical 165 genes are found
in geographically distant places (Figure 8). For example, strain TFDZ and
EML159 contain identical 16S rDNA sequences, yet the former was originally
isolated from Michigan and the latter from Oregon. It is perhaps not
surprising that the same sequence of DNA can come from places separated by
several thousand miles, since the time required for divergence in any
nucleotides may be much longer that the time for dispersal of a microbe.
This wide dispersal of identical DNA in bacteria has been observed by others,
including Sergei Bezborodnikov who found identical napthalene degrading
genes in strains from Pennsylvania, the United Kingdom and the Ukraine.

Although this study includes a dendrogram based on partial sequence
from 16S rDNA, this does not necessarily reflect the true phylogeny of these
strains, but rather only the phylogeny of the 165 rDNA segments contained
within these strains. 16S rDNA sequence analysis has become a useful tool
for understanding the origin and evolution of 16S rDNA molecule, but it
must be remembered that it accounts for only less than 0.2 percent of an
entire typical bacterial genome. However, this 16S rDNA data when used
with other information is important in understanding the evolution of these
2,4-D degrading bacteria.

Remembering Dobzhansky's words which opened this chapter we find
that knowing the evolution of 168 rDNA genes from these 2,4-D degrading
bacteria can shed light on the biology of these strains. This knowledge

92

5 map.

5. Note: M1, IMP134, and BRI6001 actuallv off thi

'J)
.9
be
c:
(C
..
U
Q)
L-
0)
f:
E
:13
Q)
U
c
‘2
:—J
cr
0)
U)
m
\C
.—
T:
U
.5
c
0)
E
U)
Q)
E
(U
.2
1:
E
l—i
2
0
U
Q)
E
M
an
as
cu
C
Q)
on
m
\D
F"
H—
o
c
o
.5
:5
..o
E
..‘L‘
'0
.2
.r:
o.
M
Li
i“
(9
co'
9
3
.9
LL

indicate strains with unique 165 sequence

 

93

provides a framework from which to evaluate the 2,4-D degradative pathways
of these strains and whether gene exchange has occurred between them,

topics which will be addressed in the following chapters.

ACKNOWLEDGEMENTS

I wish to thank Dr. Olga Maltseva for use of strain I-18 and for her
assistance in sequencing the 16S rDN A sequence from that strain. I also
thank Dr. James Cole, Amy Foxworthy, Dr. Arturo Massol-Deya, and Dr.
Jizhong Zhou for the primers and information about sequencing. I also
thank Dr. Tom Schmidt, Dr. Bonnie Bratina and Dr. Jim Smith for advice and
assistance with sequence analysis. I also thank the people in the Research on
Microbial Evolution Laboratory for discussion. This work was supported by
the National Science Foundation grant number BIR912006.

Chapter Three

Diversity of 2,4-Dichlorophenoxyacetate Degrading Bacteria Determined by Gene
Probe Hybridization Analysis

94

Diversity of 2,4-Dichlorophenoxyacetate Degrading Bacteria
Determined by Gene Probe Hybridization Analysis

SUMMARY

Many species of 2,4-D degrading organisms exist yet the genetic
diversity of the 2,4-D pathways for these strains is not known. The diversity
of 32 strains of 2,4-D degrading bacteria isolated from seven different sites in
the United States, Canada, French Polynesia and Australia were analyzed
using gene probes made from internal DNA fragments of the first three
genes in the 2,4-D pathway from Alcaligenes eutrophus JMP134, and using
the tfdA gene isolated from strain RASC, which is 73% similar in sequence to
tfdA from IMP134. Based on hybridization signal intensities under
conditions of high, medium, and low stringencies using tfdA, B and C from
pIP4, eight hybridization classes were found. On the basis of restriction length
polymorphisms, fourteen unique classes could be distinguished.
Hybridization patterns suggest that some strains contained pathways that are
mosaics of tfd genes recruited from different pathways, a finding that suggests
gene transfer and recombination. Nearly identical tfd genes were sometimes
found in phylogenetically diverse host strains, indicating that gene transfer
had occurred. A phylogenetic barrier appeared to separate two major types of
2,4-D degrading bacteria since the tfdA and tde genes from pJP4 hybridized
only to the beta and gamma Proteobacteria, but not to members of the alpha

Proteobacteria.

95

96

INTRODUCTION

If we understand the diversity of genes involved in the degradation of
2,4-D, then perhaps we will better understand how these individual genes
and collective pathways evolved.

A broad diversity of organisms capable of using 2,4-D has been isolated
(Alexander et a1. 1969; Don and Pemberton, 1981; Amy et a1. 1985; Ausmees
and Heinaru, 1990; Matheson et a1. 1991; Bhat et a1. 1994; Ka et al. 1994; Tonso
et al. 1994). Several studies have attempted to systematically characterize
these strains to gain an insight into the extent of their diversity and elucidate
patterns creating this diversity. Most of these studies have used classical I
strain characterization techniques, FAME, Biolog or REP PCR for assessing the
phenotypic and genotypic diversity of the strains. These techniques, however,
reveal nothing of a strain's genetic constitution for degrading 2,4-D.

The genetic similarities of 2,4-D degradation genes can be deduced from
DNA sequences, from amino acid sequences, or by hybridization analysis
using gene probes. The advantage of the former two techniques is that the
data are relatively unambiguous as compared to hybridization analysis. The
disadvantage is that these techniques are time consuming and costly, whereas
many strains may be quickly analyzed using hybridizations studies.

Hybridization studies of 2,4-D degrading genes that appear after 2,4-D
addition to soil have been performed using community DNA extracted from
soil (Holben et al., 1992; and Ka et al., 1994). Ka et a1. examined 47 strains of
2,4-D degrading isolates from Michigan agricultural soil that had been treated

97

with different concentrations of 2,4-D. They performed hybridization
analyses under high stringency conditions using the tfdA, tde, tde genes
from pJP4 as well as a probe (spa) that tracked a 2,4—D degrading strain of
Sphingomonas which did not hybridize to the tfd genes from IMP134. Using
these techniques, Ka et al. found the 2,4-D degraders could be classified into
four groups: 1) those that hybridized to the tfdA, tde, tde probes; 2) those
that hybridized only to the tfdA probe; 3) those that hybridized only to the spa
probe; and 4) those that hybridized to none of the probes. However, because
they performed their hybridizations under conditions of high stringency, they
may have missed some strains that have less, but still significant, similarity to
NM

This study builds upon that of Ka et al., and uses hybridizations
performed at high, medium and low stringency to detect genes which are
more distantly related to the 2,4-D genes known so far. We also used strains
from more geographically diverse regions, including Australia, Oregon,
Saskatchewan, Ontario, Montana, Polynesia, as well as Michigan. We used
five gene probes: three from the first three steps of 2,4-D degradation in
IMP134, including tfdA, tde, and tde from pJP4; the tfdA probe from strain
RASC which contains a gene 73% similar to the tfdA probe from IMP134; and
a probe which included genes of the 3-chlorobenzoate pathway from pAC27,
which has 60% sequence similarity to the lower pathway of 2,4-D degradation
in pJP4 (van der Meer, 1992).

98

MATERIALS AND METHODS
Medium and growth conditions: 2,4-D MMO containing 500 ppm 2,4-D
(Sigma) in a mineral salts medium (Stanier et a1. 1966) was used to grow the
2,4-D-degrading strains. All cultures were grown at 30°C. Broth cultures were
aerated by shaking at 200 rpm.
Bacterial strains. Strains used in this study are listed in Table 1. Strains
RASC, EML148, EML159, and EML 146 were isolated from Oregon (Amy et a1.
1985). Alcaligenes eutrophus IMP134(p]P4) (Don and Pemberton, 1981) was
provided by Dr. Ron Olson. Strains with prefix TFD were isolated from
Hickory Corners, Michigan, and Saskatchewan soils (Holben et a1. 1993; Tonso
et al. 1994). Strain BRI6001 was isolated from Nova Scotian peat (Greer et al.
1990). Strain E2w1 was isolated from forest soil in Puerto Rico, and those
strains with prefix B6 were isolated from Quebec by Dr. Roberta Fulthorpe.
Strain 1443 and K712 were isolated from Hickory Corners, Michigan (Ka et a1.
1994). Strain M1 is an oligotrophic 2,4-D degrader isolated by Dr. Fulthorpe
from the Island of Moorea, French Polynesia. Strain 2,4-D1 was isolated by
Rob Sanford from Montana.
2,4-D Biodegradation. Disappearance of 2,4-D was monitored by HPLC.
Cultures of 2,4-D medium were centrifuged at 14,000 rpm and filtered
through 0.2 u filters. These were then analyzed by HPLC on a Lichrosorb Rp-
18 column (Anspec Co., Ann Arbor) using 60% methanol and 40% 0.1%
H3PO4 as eluant. 2,4-D was detected with a deuterium lamp (Hewlett Packard

Series 1050) set for detecting wavelengths of 230 nm.

99

DNA preparation and restriction digestion. DNA was isolated using the
procedure of Ausubel (1992). DNA was digested according to manufacturer's
directions using ECORI from Boehringer Mannheim (Indianapolis, IN). DNA
molecular weight markers included bacteriophage lambda digested with

H indIII (Boehringer Mannheim). After separation by gel electrophoresis,
DNA was transferred to a nylon membrane (Amersham Hybond-NTM) and

covalently linked to the membrane with ultraviolet radiation using
StratalinkerTM 1800 (Stratagene, La Iolla, CA).

Hybridization. DNA probes were made from internal fragments of the tfdA,
tde, and 1;de genes from plasmid pIP4 (Holben, et a1, 1993). The clc probe
was made from a fragment of the the 3-chlorobenzoate degrading plasmid,
pAC27. The tfdA probe from strain RASC (here called rscA to differentiate it
from tfdA from pJP4) was provided courtesy of Dr. Yuichi Suwa. Plasmid
vectors containing the genes were digested with appropriate restriction
enzymes, purified by gel electrophoresis, excised from agarose, and labeled
with digoxygenin-l 1-dUTP from (Boehringer Mannheim ) using the Random
Primers Labeling Kit (Boehringer Mannheim) according to manufacturer's
instructions. All hybridizations were carried out at 62°C. High, medium, and
low stringencies were achieved by varying formamide concentrations in the
hybridization buffer from 50%, 20% and 0%, respectively (Wahl et a1. 1987).
After hybridization, blots were washed in double strength wash buffer at
room temperature for 5 min, then subjected to colorimetric detection using
alkaline phosphatase with X-phosphate as substrate according to
manufacturer's directions. Photographs of blots were made using Polaroid 55

positive negative film.

100

RESULTS

The diversity of 2,4-D genes from 32 bacterial strains of 2,4—D degraders
(Table 1) was assessed using gene probe analysis. DNA from each of these
strains was isolated and digested with EcoRI, separated by gel electrophoresis,
and transferred to a nylon membrane. We used the technique of Wahl et a1.
(1987) for optimizing hybridization conditions for detecting genes with
sequence similarities in the ranges of high (90-100%); medium (75-90%) and
low(55-75%) sequence similarities (Figure 1).

The calculated similarity required for hybridization was confirmed
experimentally by finding that the tfdA probe from pJP4 hybridized to strain
RASC only under conditions of low stringency, which agrees with the 73%
sequence similarity found for that q‘d gene in this strain.

Nylon membranes containing DNA from each of the strains were
hybridized at high, medium or low stringencies to each of the probes, tfdA,
tde, tde, clc, and the rscA probes (Table 2). The tfdA probe from RASC was
used to hybridize the blots under low stringency conditions only.

These strains could be grouped in two ways: either on the basis of
hybridization stringency when hybridized with the probes, which indicates
degree of similarity of the genes, or a more refined grouping based on
restriction patterns, which may indicate a gene's similarity and location in the
host genome. When restriction fragment sizes were considered along with
the hybridization stringency patterns, at least 14 distinct 2,4-D gene patterns
could be distinguished (Table 2). When grouped solely on the basis of
hybridization intensity using the tfdA, B and C probes, eight unique groups

. 28.8th mm: 8838888 583 8883888 888:8: ..D.. .5889: mmm 8:882 £83 «£88.88 2 8898 go mm“ 8° 83:58: oEmm ...

AN 888:8 88.5 <20: 88 8o 88: 5 88588 258m ..

 

101

 

 

D 8 53—: 8.58:3: mucosonaoumcnfim 2805—88..— stoaom 385.0: 588.8: 82
a 8. 288 .8 85588.88 82 ..8: s .88 58:82 88.88
D 88 2&5 in muroEomzimm 8.82 ....8 as $6.88: :omBO 332m
D v 53—: an m§sE°m=mfim 09858:"— 283880 3.3
D 88 29a in mugs—8.38:3 0905—888— 3825 miom
88 8. £888: .8 8558.88.88 285m 5:38 «5.52 5.8..N
D 8 33 «835.5% antxenoxx 28.85.88: 2.82880 8.3
8 8 s»: 88558 8:583:88 82 .8: s 8:3 52558888 893
N 82 .8: s 8:3 5:88:82 83
N 82 ...: s 8:3 55:85.88 29:
N 82 ...: s 8:8. 58:82 898.8
N N 88 .8 888925: 82 .8: s 8:3 58:82 N93
8 82 .8: s 8:3 58:82 89:.
8 82 ...: s 8:8. 58:82 89:
8 82 ...: s 8:3 5:88:82 8.893
8 82 ...: 8... 8:3 5:88:82 88.88.
8 82 .8: 8... 8:3 888.5 29:.
8 82 .8: 8... 8:8. 5:88:82 29:.
8 82 ...: 8 8:3 58:82 889:.
8 N as: .8 88288.8 82 .5 8 8:3 58:82 293
8 «89888:: 38388 8888 2:28: 838
8 N as: .8 8.58.8225: 82 520 58:82 88888:
8 N s»: .8 858935: 82 ...: s 8:3 58:82 89:
8 N 38 .8 853885.: 82 ...: s 8:3 58:82 :9:
88 N 88 .8 8.888588 82 .8: s .85.: 58.0 098
an N San unscugx nukxotgx mom" ...: 8a 098,—. 8398—32 ”NOE.
m N 83 8:839:88» 85288;: man J: «0 02.8.8. 8:85:82 wan—"E.
m N :8: 8.88:... 8888.8: 82 .8. 8... 8:3 58:82 889:
D N 83 8838338 85%.:qu mam ...¢ 5 388,—. 532388—885 an»:
88 N :8: .8 88.58385: 82 ...: 8... 68 58:82 28
88 N as: .8 858288588 82 ..8. 8... 8:3 55:85.88 8:3
88 . 82 .8: 5 .885. 58:0 88.8288
88 N s»: .8 85.8935: 82 .8: s 8:2 5:80 88.8288
88 N as: 8.88:8 88888:: 888 £28.88 :5 :88 88889... 388.828
E88888 8288.888 :3 5886888 882 8588.88 58833 .53.
858.8882: 8 838
88—858» mm»— 38 32898398 655.3 .885 £538: wcmvauwuv D..¢:N .N «Ssh.

102

 

..........................
oooooooooooooooooooooo

535533353; 33:35:}: 5:333:55 35:32:; 3:32:35: Total Genomic

nnnnnnnnnnnnnnnnnnnnnnnnnnn
oooooooooooooooooooooooooo

...........................
..........................

...........................
..........................

$333232; izézézizézé zézézézégi; 22:22: 52522223: DigeStiO"

...........................
..........................

{5:55;}: 5:33:33; 23:23:32}: 3:33:33: 313535333? With ECORI

..........................
...........................
..........................
...........................
...........................
..........................
...........................
..........................
...........................
..........................
...........................
..........................
...........................
..........................
...........................
..........................
---------------------------
uuuuuuuuuuuuuuuuuuuuuuuuuu
...........................
..........................
...........................
--------------------------
...........................
oooooooooooooooooooooooooo
...........................
..........................

 

 

..................................................... High Stringency

..........................
---------------------------
--------------------------
ooooooooooooooooooooooooooo
..........................
uuuuuuuuuuuuuuuuuuuuuuuuuuu

332225;? 35532523; igigégégégé 52:53??? 25355525? Hy brdization

..........................
uuuuuuuuuuuuuuuuuuuuuuuuuuu

---------------------------
oooooooooooooooooooooooooo
ooooooooooooooooooooooooooo
..........................
..........................
..........................

5355555353 53555555593 5553555535? 5323??? iii-35%??? Strain A > 90% Similar

----------------
..........................
...........................

uuuuuuuuuuuuuuuuuuuuuuuuuuu
uuuuuuuuuuuuuuuuuuuuuuuuuuu
oooooooooooooooooooooooooo
nnnnnnnnnnnnnnnnnnnnnnnnnnn
..........................
...........................
..........................
ooooooooooooooooooooooooooo
--------------------------
---------------------------
..........................
...........................
..........................
---------------------------
uuuuuuuuuuuuuuuuuuuuuuuuuu
ooooooooooooooooooooooooooo
---------------------------
aaaaaaaaaaaaaaaaaaaaaaaaaaa

 

 

Medium Stringency

 

 

Hybrdization

.... . . .
3333323333 5533325532; 2332325233 332323332 525323533; Strains B and C 75 - 90% Similar
.... ......

A B C D E LOW Stringency

...... Hybrdization

-

_ éiiiiiiiii Strain E 50 " 75% Similar

'I‘I-I-I-Z 'I-Z-I'Z'I' um 'I'I'I-I-Z '2-3-2-2'2- . . .

5:222:22: 2:32:23; .... 22:22:22 2:222:32; Strain D < 60% Similar

uuuuuuuuuuuuuuuuuuuuuuuuuu
ccccccccccccccccccccccccccc
nnnnnnnnnnnnnnnnnnnnnnnnnn
...........................
oooooooooooooooooooooooooo
...........................
oooooooooooooooooooooooooo
...........................
oooooooooooooooooooooooooo
ooooooooooooooooooooooooooo
oooooooooooooooooooooooooo
...........................

 

 

 

Figure 1. Strategy for hybridization at high, medium, and low
stringencies.

103

Table 2. Hybridization patterns of 2,4-D degraders. N umber(s) indicate restriction
fragment sizes (sizes indicated only once for each group of strains). Fragments
marked with *“ hybridize at high stringency, those "" at medium stringency and

those with * at low stringency. The symbol "a" means no hybridization.

 

EcoRI Fragment Sizes

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Strain Location REP" Subgroup tfdA rscA tde tde clc

JMP134 Australia U beta 2 12‘" 12"“ 16"" 16"“ 16""

EML159 Oregon U beta 2 8"" 8’" 8"“

EML157 Oregon U rd

TFD39 Saskatchewan U beta 2

K712 Michigan U beta 2 6.3"" 3.5"“ 3.5“ 6" 6*

TFD9 Saskatchewan; U beta 2 3.5""

TFD41 MiChigan 5 hem 2 4.5.5. 4.51.! 20’ 16"." 20... 6+

TFD38 Micflgan 5 beta 2 2.6"" 4.1"" 16, 6'“

TFDZB Michim 7a beg 2 7.2, 3.5”" rd 3.5’ 6.3‘ a

TFD31 Saskatchewan 7b beta 1 18* 14* 9 a a

86-9 Ontario U beta 1 20, 15” 20, 15* a 20" 20""
2.3”

TFD6 Michigan 1 beta 2 9.5“ 9.5““ 9.5'" 9.5"" 9.5"

TFDZO Michigan 1 beta 2

BRI6001 Quebec 1 beta 2

TFDI3 Michigan 1 beta 2

TFD6 Michigan 1 rd

TFD18 Michigan 1 rd

TFD17 Oregon 1 rd

TFDZI Michigan 1 rd

TFDI4 Michigan 1 rd

TFDIS Michigan 1 rd

TFD27 Michigan 1 rd

Ezwl Puerto Rico 1 rd 11.3" 9.5"" 9.5“" 9.5"" a

RASC Oggon U beta 2 7"‘ 7’” 7, 3.5" a a

TFDZ Michigan 2 beta 2 9.5” 9.5" 9.5" a a

TFD4 Michigan 2 rd

TFD18 Saskatchewan 2 rd

TFD7 Michigan 2 rd

2,4-D1 Montana U alpha 4 9 rd > 23, 8" a a

EML146 Om U a_lpha 4

86-5 Ontario U alpha 4 a nd a a a

86-10 Ontgio U alpha 4

K1443 Michigg U alga 4 8"

M1 French Polmeaia U alpha 1 a rd 12" a o

 

 

" REP group designation of each strain is indicated by number or by "U" which

indicates unique REP group (Fulthorpe et al., submitted).

104

could be distinguished (Figure 2). Three of the strains did not hybridize to
any of the probes.

Different tfd genes: Differences in hybridization stringencies of gene probes
gives us an insight into differences in DNA sequences for these genes. For
example, in the strains examined here with the tfdA probe from pJP4, three
classes of gene sequences could be distinguished One class, hybridizing at
high stringency, included strains IMP134, EML159, TFD39, TFD41, TFD 38,
K712, TFD9 and TFD23. Another class of tfdA strains included TFD2, RASC,
E2w1, TFD31, 86-9 and TFD6 and their siblings. These strains hybridized to
tfdA from pJP4 only under low stringency conditions, indicating a theoretical
sequence similarity between 55-73% (Wahl et a1. 1987). Actual sequence data
for the tfdA genes from strains TFD2, RASC, E2w1 and TFD6 showed that
they are identical to each other and 73% similar to tfdA from pJP4 (chapter 4).
The third class of strains, including strains EML148, 2,4-D1, 86-5, 86-10, K1443,
and M1 did not hybridize to tfdA from pJP4 at all. None of the strains
hybridized at medium stringency to tfdA.

On the basis of the RFLP patterns observed when probed with the tfdA
gene probe, there were 10 distinct types of tfdA - containing strains. Those
strains which had identical restriction fragment length polymorphisms, such
as TFD6, TFD14, and TFD15 were assigned the same RFLP class designation.
Not surprisingly, most strains produced the same restriction pattern when
probed with tfdA from either pIP4 or from strain RASC.

Four types of tde genes could be resolved on the basis of hybridization
signals: those that hybridized at high, medium, low, and not at all to tde.
However, differences in RFLP patterns indicated 10 unique types of tde

105

tfdA :de :de

JMP134, TFD41

K712, TFD9

TFD23

TFD31, B6-9

TFD6, TFDZO

RASC, TFD2

2,4-D1, M1

86-5, 86-10

 

Figure 2. Hybridization patterns with tfdA, B and C suggest a mosaic
structure of these 2,4-D degrading strains. Solid boxes indicate high
similarity to pJP4; striped. boxes indicate medium stringency; stippled
boxes indicate low stringency; and white boxes indicate that no
hybridization was observed.

106

classes. A tde-like element was phylogenetically the most widely distributed
type of tfd gene in the strains probed.

Three types of tde genes could be resolved on the basis of
hybridization intensity: those that hybridized at high stringency, at low
stringency and not at all to q‘dC. By RFLP pattern differences, seven unique
types of q’dC classes could be determined.

All but four of the strains used in this study hybridized to at least one
of the probes used in this study. Only strains M1, 36-5, and B6-10 hybridized
to none of the probes, indicating 2,4-D degradation gene sequences different
from those of IMP134.

The hybridization patterns clearly showed that nearly identical tfdA
genes are paired with different tde and tde genes in different strains (Figure
2). This mosaic structure of tfd genes can beclearly seen in strains IMP134,
strain TFD6 and strain RASC. In this example, the tfdA gene in RASC and
TFD6 are identical and different from IMP134. Interestingly, hybridization
analysis shows that strains IMP134 and TFD6 share identical 1;de and C genes,
which are different in strain RASC.

Comparison to 16$ Phylogeny: Representatives of unique strains were
compared to a phylogeny (chapter 2) of these strains made on the basis of 165
sequence comparisons (Figure 3). This figure illustrates that tfdA and C genes
similar to those from pIP4 are found only in the beta and gamma
Proteobacteria. Furthermore, the tde gene was detected at low stringency in
only three of the alpha Proteobacteria .

1 07
tfdA tde tde

B. cepaciaDS
B. cepaa'afA
TFD39

B. mailer . . .
B. gladioli
BR16001
TFD6
EML159
TFD2

B. caryrphylli
B. andmpogmis
RASC

K712
IMP134
TFD41
TFD38

B. solarxmm
A. eulnphus

A. Wars
TFD9
"D23
TFD31

369

R. Wars
C. tadaslostemnr’
H. elongate

D. halaphfla

148 O
O

 

i
i

@QO.

Exolr'

24D1

5. Mac
5. Mudflats
F. capsdatum
865

3610
K1443
EML146

5. My:
R. palastn's

M1 Q ~@

D. lid/'1?

@000 so

 

Figure 3. Hybridization patterns of tfd A,B and C probes from pJP4 to
2,4-D strains (in bold type) arranged in phylogenetic grouping (branch
lengths not to scale). Solid dots indicate high similarity; striped dots
indicate low similarity; stippled dots indicate low similarity; and open
circles indicate no hybridization. Solid branches are members of the beta
Proteobacterrh, striped are gamma Proteobacteria stippled are alpha
Prateobacteria and the lone solid branch is delta Proteobacten'a.

108

DISCUSSION

This study showed a greater diversity of 2,4-D degradation genotypes
than has previously been reported. This diversity is partly due to 2,4-D
degrading genes assembled with other 2,4—D genes in different combinations
in different strains. This assembly if identical genes in novel arrangements
has created a mosaic of genes in the 2,4-D degradation pathway. It appears
that tfdA and tde genes similar those of strain IMP134 are phylogenetically
limited to the beta and gamma Proteobacteria.

Diversity: High, medium, and low hybridization stringencies can detect a
greater diversity of 2,4-D degrading genes which may be missed by high
stringency hybridization alone. For example, using high stringency
conditions alone, tfd genes in strains RASC, TFD2 and others would not be
detected. Using a strategy of hybridizing at three different stringencies with
tfdA, tde and tde, we found eight hybridization classes of 2,4-D degraders,
representing a greater diversity of 2,4-D gene types than the four classes found
by Ka et al. 1994.

We also found that the diversity of tfd hybridization patterns of these
strains correlates fairly well to the diversity of REP classes (Fulthorpe,
personal communication; Tonso et a. 1994). In all cases but one, unique REP
classes gave unique hybridization classes. The one exception was E2w1,
isolated from Puerto Rico, which belongs to REP class 1 along with 10 other
strains (including TFD6) used in this study (Fulthorpe, personal
communication; Tonso et a1. 1994), but which has a hybridization pattern

different from the others. The different fragment sizes containing tfdA in

109

strains TFD6 and E2w1 may be explained either by gene rearrangement in the
host or by a point mutation resulting in the loss of an EcoRI site.

Mosaic pathways: The results presented in this study indicate mixing of
genes in the 2,4-D pathway between strains, which supports the notion that
evolution has occurred by a process of molecular "tinkering" in which bits
and pieces of DNA are spliced into new functional units (Jacob, 1977). For
example, some strains such as TFD23, K712 and TFD9 contain tfdA genes with
greater than 99% similarity to IMP134, yet contain genes for the third step in
the pathway which are less than 60% similar to the tde gene from JMP134.
Strain B6-9 also appears to have a gene that is > 90% similar to cch, whereas
its close relative, TFD31 does not.

Various pathways for 2,4-D degradation may have arisen by the splicing
together of different genes in the pathway. For example, some tfdA genes
highly similar to those in IMP134 are combined with low stringency tde
genes, as is the case with TFD9 and K712. Another example is that tde and C
genes highly similar to those of pIP4 are present in TFD6, but not in strain
RASC, which appears to have a tfdA gene identical to that of TFD6, which has
been shown to be the case (chapter 4). Among the strains that hybridize to
tfdA at high stringency, strains TFD19, K712, and TFD 41 hybridize to tde ,
whereas 86-9 and TFD31 do not. Among the strains with tfdA genes nearly
identical to strain RASC, only strains with RFLP patterns identical to TFD6
and E2w1, have tde genes nearly identical to IMP134, whereas strains RASC
and TFD2 do not. The apparent rearrangements become even more mixed:
strains TFD6 and F.2w1 contain tde genes highly similar to those of pJP4,
whereas RASC and TFD2 do not hybridize to this gene probe at all. It is as if

110

TFD6 and related strains evolved from the recombination of tfdA from a
RASC-like strain and tde and C from a IMP134-like strain (Figure 4).

The recombination of tfdA , tde, and tde suggests that some of these
genes may be in separate operons, as has been shown for the tfdA, the tde
and the tdeDEP operons of JMP134. This may be the case for those strains
which hybridize to plasmid pJP4, including TFD41, K712, and T'FD23. In these
strains, the tfdA and tde genes are on different EcoRI fragments. The tde
gene appears to be more closely situated to either the tfdA, in the case of K712,
or to tde, in the case of TFD41 and IMP134. Others strains, in contrast, appear
to have genes for the whole pathway in close proximity. For example, in
strain RASC, the tfdA and tde genes are located on the same 7.0 kb EcoRI
fragment, which is indeed the case as confirmed by Suwa et a1. (1994) and
Kamagata (in preparation) in sequence analysis of the tfdA and tde genes.
Also, it appears that the entire pathway is contained within a single EcoRI
fragments in strains TFD6 and in TFD2.

Gene transfer: The recombination of genes from apparently different
sources may be explained by gene transfer. When the hybridization data
presented here is compared with the phylogeny based on comparisons of
partial 16S rDNA sequence data, it becomes apparent that similar tfd genes are
found in phylogenetically diverse strains. One explanation for this is that the
tfd genes have transferred between species.

It appears that gene transfer is limited to within deep phylogenetic
divisions, since tfdA and tde genes are found only in the beta and gamma
Proteobacteria, but not in the alpha Proteobacteria.

Similar 2,4-D genes found in different strains provide evidence of gene

111

t fdA tde t de

   

IMP134, TFD41

 

RASC, TFD2

 

 

 

TFD6, TFDZO

 

 

Figure 4. Mosaic pattern of tfd genes from three strains indicates
recombination. The solid boxes represent genes that are highly
similar to those found in IMP134 as determined by hybridization
analysis in the case of tde and tde. The stippled boxes represent
genes identical to those from RASC as determined by sequence
analysis in the case of tfdA.

112

exchange. For example, strain IMP134, TFD 19, and TFD39 represent three
unique strains based on FAME, Biolog, and REP patterns (Tonso et al. 1994;
this paper) and on 16S rDNA sequence data (chapter 2). However, all three of
these strains appear to carry identical 2,4-D genes. This is not surprising, since
plasmid pIP4 from JMP134 is a broad host range IncP1 plasmid which has
been shown to transfer into all Gram negative bacteria except Bacteroides
(Don and Pemberton, 1981; Harker et al., 1989; Friedrich et a1. 1983). These
strains differ by more than 9% in 168 rDNA sequences, suggesting that these
strains diverged more than 100 million years ago (Moran, 1993). However,
identical RFLP patterns in the 2,4-D genes of these strains which represent up
to at least 40 kb of conserved DNA sequence (cumulative sizes of hybridized
fragments) indicate that the 2,4-D genes have changed very little, probably less
than one percent (Matheson et al. 1994; Suwa et a1. 1994; and Moran et a1.
1993). Because the similarity of the genes is so high and the divergence of the
strains is so great, this provides some evidence for interspecies gene transfer
in situ. This hypothesis is treated more explicitly in the following chapter.
Different tfd genes have been found in identical strains, which further
suggests that gene exchange has occurred. Strain TFD23 and TFD31 are
identical at the level of 16S rDNA (chapter 2), yet they differ from one another
in hybridization patterns for 2,4-D genes. Because the host organisms are so
closely related, but the genes have different restriction patterns and sequence
similarities to pJP4, this is evidence of gene transfer.
Gene flow of the tfd genes may be restricted because of some

evolutionary constraints since the hybridization classes clustered into two

broad categories based on phylogeny: those that hybridize with the tfdA probe

113

and those that do not. All those which hybridized to tfdA were members of
the beta and gamma groups of Proteobacteria and those that did not were
members of the alpha subgroup of Proteobacteria. Furthermore, all but three
of these beta Proteobacteria that hybridize to tfdA are most closely related to
the genus Burkholderia (Yabuuchi et a1. 1992).

All of the strains that did not hybridize to tfdA were found to be
members of the alpha group of Proteobacteria, according to the classification
of Woese (1979) and all but one of these were found to be closely related to the
genus of Sphingomonas. Interestingly, some members of the
Sphingomonas group of 2,4-D degraders hybridized to the tde gene probe at
low stringency, indicating a gene with a sequence similarity of about 60% to
that of tde.

The one strain that did not hybridize to tfdA and was not closely
related to the genus Sphingomonas was strain M1. This oligotrophic strain
was most closely related (97%) to the photosynthetic bacterium
Rhodopseudomonas palustris, which is a member of the alpha class 2 of
Proteobacteria. Strain M1 may contain tfd genes which differ significantly
from pIP4 and from the Sphingomonas strains, though this was not
detectable in these hybridization analyses.

This work has shown that the diversity of 2,4-D pathways looks like a
mosaic of genes derived from different places. "Evolution does not produce
novelties from scratch," wrote Francois Jacob in 1977. "This paramount
reshuffling of genetic elements provides tremendous potentialities of
adaptation." It appears that much of the reshuffling genes in 2,4-D pathways

has been a result of interspecies gene transfer; however, this transfer appears

114

to be limited to within deep, phylogenetic groupings.

ACKNOWLEDGEMENTS

I thank Dr. Roberta Fulthorpe for her collaboration in the hybridization
work and for the use of her strains. I also thank Dr. William Holben for 2,4-D
strains and tfdA, 1:de and tde gene probes and Dr. Jong-Ok Ka, Dr. Nancy
Tonso, Dr. Penny Amy and Dr. Charles Greer for the use of their strains.
Thanks to Yuichi Suwa for the use of his gene probe from RASC. I also thank
the people in the Research on Microbial Evolution Laboratory for discussion.
This work was supported by the National Science Foundation grant number
BR912006 and by Japan Research and Development Corporation grant

number 1.

Chapter Four

Interspecies Gene Transfer in the Evolution of
2,4-Dichlorophenoxyacetic Acid-Degrading Bacteria
in Nature and in the Laboratory

115

SUMMARY

Many phylogenetically diverse 2,4-D degrading bacteria exist in nature,
yet it is not known how this diversity of organisms evolved. In this study,
the role of gene transfer in the evolution of this diversity was examined. To
do this, the phylogeny inferred from their 16S ribosomal DNA sequences was
compared to a dendrogram of sequences for the tfdA gene, which encodes the
first enzyme in the 2,4-D degradation pathway. DNA molecules from 2,4-D
degrading strains of Proteobacteria were amplified by PCR using primers
made from conserved regions of the previously sequenced tfdA genes from
Alcaligenes eutrophus JMP134 and strain RASC. Three classes of tfdA-like
sequences were found in these strains. One class of tfdA sequences was
identical (or nearly so) to that from IMP134 and the other two classes each
differed by 23.0% and 22.4% from JMP134 and 6.4% from each other. The tree
derived from these tfdA genes differed dramatically from the phylogeny of
the 168 DNA from the same organisms, providing evidence of interspecies
horizontal gene transfer. Furthermore, identical tfdA gene sequences were
distributed among several deep branches of the beta and gamma subgroups of

Proteobacteria. To discover whether these strains could mobilize 2,4-D genes
in vitro, they were mated with a 2,4-D‘ recipient strain, Burkholderia cepacia

D5. Several strains transferred genes which expressed the ability to degrade
2,4-D. Those that rapidly transferred and expressed the 2,4-D phenotype
included strains with tfdA sequences greater than 99% similar to tfdA from
JMP134. Those strains that did not transfer the ability to degrade 2,4-D, or did
so only at extremely low frequencies, had tfdA gene sequences that differed by
up to 23% from JMP134.

116

117

INTRODUCTION

How did the diversity of 2,4-dichlorophenoxyacetate-degrading
microorganisms evolve? This is more than an esoteric question, for to learn
how xenobiotic degraders evolved may assist in better management of
xenobiotic pollution. One hypothesis to explain the evolution of diversity is
horizontal gene transfer between species (Wilson, 1977).

Many studies have examined interspecies gene transfer (Brisson-Noel
et al., 1988; Chatterjee and Chatterjee, 1987; Datta and Hedges, 1972; Fulthorpe
et al., 1992; Mazodier et al., 1989; Natarajan and Oriel, 1991 ; and Trieu-Cout et
al., 1987) and many review articles and books have been written on this topic
(Coughter and Stewart, 1989; Fry and Day, 1989; Levy and Marshall, 1988;
Levy and Miller, 1989; Stotzky,1989; Stotzky and Babich, 1986; and Trevors et
al., 1986) yet few studies have examined gene transfer between different
phylogenetic groups to gain an understanding of the evolution of organisms
in nature (Wilson, 1977).

Horizontal gene transfer of the 2,4-D degrading plasmid pJP4 has been
documented in laboratory studies (Don and Pemberton, 1981 ; Harker et al.,
1989; Kukor et al., 1989; Friedrich, 1983), yet it is not known to what extent
other 2,4-D genes have been exchanged in nature resulting in the evolution
of new strains. The aim of this research was to assess the role of horizontal
gene transfer in the evolution of 2,4-D degrading strains.

To demonstrate that gene transfer has occurred in nature involves
comparing the phylogenies of two separate genes in a group of organisms, a

"phylogenetic congruency test" (Wilson, et al., 1977) or by determining that

118

identical genes are found in widely different strains. Within the same strains,
different genes having different phylogenies indicates that horizontal gene
transfer has occurred (Dykhuizen and Green, 1991; Wilson et al., 1977).

To test the hypothesis that gene exchange was responsible for
generating the diversity of 2,4-D degrading strains, I performed a phylogenetic
congruency test comparing 16S rDNA sequences, discussed previously
(chapter 2), with the tfdA gene sequences discussed here.

The tfdA gene was chosen for this because it is the only gene in the
2,4-D pathway known to be used exclusively in 2,4-D degradation, unlike fidB
and tde, which have homologs found in other degradative pathways. The
tfdA gene encodes an alpha-ketoglutarate-dependent 2,4-D dioxygenase
(Fukumori and Hausinger, 1993) and is the first enzyme in the 2,4—D pathway
which converts 2,4-D into 2,4-dichlorophenol and glyoxylate. This 861 bp
gene, first sequenced from Alcaligenes eutrophus JMP134 (Streber et al., 1987)
has identical sequences to those of strains EML159 and TFD38, (Fukumori,
personal communication) and TFD41 (Nakatsu, personal communication).
(TFD38 is identical to strain TFD41 by REP-PCR, by fatty acid methyl ester
analysis and substrate utilization profiles (Tonso, et al., 1995) and by
hybridization analysis (chapter 3)). These tfdA sequences differ by 25% from
Burkholderia strain RASC (Suwa, 1994) and Burkholderia strain TFD6
(Matheson, 1994) which contain identical tfdA sequences.

Although the tfdA gene of Alcaligenes eutrophus IMP134 is on a
plasmid, other tfdA genes appear to be chromosomally located, including
RASC (chapter 5; Suwa, et al, in prep); TFD6 (Matheson, et al., 1994); and
BRI6001 (Greer, et al., 1990) which appears to be identical to TFD6 (chapter 3).

119

Furthermore, Ka et a1. (1994) found that 25% of the tfdA hybridizing strains in
his collection appeared to have this gene located on the chromosome.

The 2,4-D degraders used in this study belong to the alpha, beta and
gamma subgroups of Proteobacteria, a diverse group of Gram negative
bacteria (Murray et al., 1990). The beta subgroup of 2,4-D degraders includes
strains related to Alcaligenes eutrophus, Burkholderia cepacia, Alcaligenes
xylosoxidans, and Rhodoferax fermentans. A single member of the gamma
Proteobacteria most closely related to the genus Halomonas was also
examined. An additional strain used in this study clustered with the genus
Sphingomonas, a member of the alpha Proteobacteria. It was found that all
strains belonging to the beta and gamma subgroup of Proteobacteria
hybridized to tfdA, whereas none of the strains clustering with the alpha
Proteobacte'ria hybridized to tfdA (chapter 3; Fulthorpe, et al., 1995).

In this chapter, I describe experiments designed to test the hypothesis
that the genetic diversity of 2,4-D degrading organisms is due, in part, to
interspecies gene transfer. I examined the phylogenies of the tfdA and 16S
rDNA genes from a group of 2,4-D degrading strains, and found that their
phylogenies were different, indicating that gene exchange had occurred in
nature. To further test the hypothesis that gene transfer occurs among this
group of 2,4-D degrading strains, in vitro mating experiments were
conducted with these strains. These experiments showed that 2,4-D genes
were transferred at relatively high frequencies from strains which hybridized
to plasmid pJP4 and at very low to undetectable frequencies in those strains

which did not hybridize to plasmid pJP4.

120

MATERIALS AND METHODS

Media and growth conditions: 2,4-D degrading strains were grown in 2,4-D
MMO containing 500 ppm 2,4-D (Sigma) in a mineral salts medium (Stanier,
et al., 1966). When appropriate, antibiotics including kanamycin, carbenicillin,
and bacitracin, were added at concentrations of 50 ug / ml. PTYG contained
0.25 g peptone, 0.25 g tryptone, 0.5 g yeast extract and 0.5 g glucose, 0.03 g
magnesium sulfate, and 0.003 g calcium chloride per liter. Trypticase soy
broth (TSB) was obtained from Difco (Detroit, MI). All cultures were grown
at 30°C. Broth cultures were aerated by shaking on Innova 4300 shaker (New
Brunswick Scientific Co., New Brunswick, NJ) at 200 rpm.

Bacterial strains: The strains used in the phylogenetic congruency tests are
described in chapter 2 and listed in Table 1. Additional 2,4-D degrading strains
used in mating experiments are listed in Table 2. All of the 2,4-D degrading
strains are sensitive to 50 pg / ml each of the antibiotics kanamycin,
carbenicillin, and bacitracin. Burkholderia cepacia D5, which contains
neomycin phosphotransferase genes carried on Tn5, was supplied by Ron
Olsen (Walch and Ballou, 1983) and is resistant to 50 ug/ ml each of
kanamycin, bacitracin and carbenicillin.

Sequencing of tfdA-like genes: Conserved primers internal to tfdA,
designated TVL and TVR, were prepared by Tatiana Vallaeys and Alice
Wright (Wright et al., in prep). These primers were used to amplify
sequences by PCR from the following organisms: TFD2, B6-9, K712, TFD9,
TFD39, TFD31, TFD23, I-18, and K1443. Negative controls (water) were used to
insure no contamination of PCR products. Amplified DNA fragments were

121

size fractionated on a 1.5% electrophoresis gel. Primers were removed from
the PCR amplified DNA using the GeneClean Kit (BiolOl, La Jolla,
California). Approximately 20 pmol of primer TVL was added to 150 ng
amplified DNA in 20ul total reaction volume. DNA was sequenced using the
Applied Biosystems Model 373A automated sequencer (Perkin Elmer Cetus)
at Michigan State University Sequencing Facility.

Mating studies: A ten-fold excess of recipient strains B. cepacia D5 in broth
(~ 109 cells) was mixed with donor cells and centrifuged briefly to concentrate.

Mixed cells were then pipetted either onto a 0.2 urn filter, which was placed
onto nutrient agar plates, or were pipetted directly onto selective agar media.
Cells were allowed to incubate together overnight and portions were then

resuspended in H20 and serially diluted before plating, and other portions

were streaked directly onto selective plates of 2,4-D kanamycin, carbenicillin
and bacitracin. IMP134 was mated with B. cepacia D5 as a positive control.
The recipient alone and the 2,4-D degrading strains alone were carried
through the mating procedures as negative controls.

2,4-D biodegradation: Disappearance of 2,4-D was monitored by HPLC.
Supernatants were centrifuged at 14,000 rpm and filtered through 0.2 u filters
and were then analyzed by HPLC on Lichrosorb Rp-18 column (Anspec Co.,
Ann Arbor) using 60% methanol and 40% percent 0.1% H3PO4 as eluant.

2,4-D was detected at 230 nm.

Hybridization: DNA from the entire pJP4 plasmid and the nptII gene of Tn5
were labeled using random hexanucleotides incorporating (oo-32P)dCTP (3,000
Ci/mmol; New England Nuclear, Boston, MA). Hybridizations were done

122

under low stringency conditions using 50% formamide and Denhardts
solution (Maniatis, et al., 1985) at 42°C.
RESULTS

Strains: The strains chosen for comparison of 168 and tfdA DNA sequences
are listed in Table 1. With the exception of strain K1443, which contains a 165
rRN A sequence belonging to the alpha Proteobacteria and which did not
hybridize to tfdA, all strains are members of the beta and gamma
Proteobacteria (chapter 2) and all hybridize to tfdA from pJP4 at either high,
medium or low stringency (chapter 3). All strains were shown to be unique
either by 16S rDNA sequences, by unique repetitive extragenetic palindromic
(REP) PCR patterns (Fulthorpe et al., 1995), and / or by unique hybridization
patterns with probes from internal regions of the tfdA, tde, or tde genes
encoding the first three reactions in the 2,4-D pathway (chapter 3). These
strains were isolated from diverse locations in Australia, Oregon,
Saskatchewan, Ontario, and Michigan.
Sequencing of tfdA-like genes: Conserved primers internal to the tfdA gene
were used to amplify DNA from strains TFD2, B6-9, K712, TFD9, TFD39,
TFD23, TFD31, I-18, and K1443. These amplified DNA fragments were
approximately 320 nucleotides in length as determined by size fractionation
on a 1.5% electrophoresis gel. This is the same size as the tfdA gene fragment
from pJP4 and strain RASC.

Sequences of tfdA DNA were initially aligned in Sequ (ABI) and were
imported into MacClade for further analysis. Additional tfdA sequences from
J'MPlB4 (GenBank accession number M16730), from strain RASC (courtesy

123

856636 8: was 8:34.33 as: 8.8%,: as... 98

3:88 accesses: o: messes < 6:855 Ema 3 ER 35% s 853.3 can» afi asses ..+.. < ..

.uxB 5 :03» fl modem <35 mo #8335933 ._

 

 

v «in? .mm meeosomfixnm - 33 >130 «V— e dummy—32 9.3M

«8:30 35% 2:85.03: 0: mag :3 “0 «$8sz E ~8qu EA

H 30m menaesntmx Heaukonocm p: mag :3 .0 09053..“ E 3330 mém

H 30m mneuemfitx nuamxoncxm - mad“ :3 no omaop. 5 5305.33 awn—E.

N 33 as assessed - $2 ..a a cash. a cameos: SE

N sum as assesses - we: ..a a cash a cafe: 00E.

N sum a... assessed - $2 ..a a .82 a 555 092

N San 335.50% Ruaukcuoxm + mmofi :1 «o oncoh _ cmwfiowz mNDmH

N Sum 8233:.» mucumzeoa + mag in no 098,—. u :meoaz HE

N San maeumxomoix Bahamas; + mmmfi :3 so 09.5% H 53052833 8%.

N .28 a... assessed + 32 ..a a Q _ 595:2 S?

N 20m A? Sausofitam + mama :3 no 8:8. m 53052938 $03.

N son a... etaezofitsm + 33 .a a .32 H 830 $35

N San 2:323» mscmmmuui + $3 dotonfimm was con— H «39:35. 35:):

magma—am . 3h. damn—mm .32 .333 mucouuuox anus—U 5530..— £35
313030895 Sub

 

63:0st <ZDu may 65 <-b no saggy—8 now won: mama—um A 033—.

124

of Alice Wright and Yuichi Suwa), from EML159 (courtesy of Fumiyasu
Fukumori) and from TFD41 (courtesy of Cindy Nakatsu) were added to the
alignment (Figure 1). Three classes of tfdA gene sequences were found: class I
included IMP134, EML159, TFD41 and strain TFD39 which differed by only
one nucleotide from JMP134, and strains TFD23, K712, and TFD9, which
differed from IMP134 by 1 base pair. The class II tfdA sequence was identical to
that found in RASC and TFD6 and included TFD2. The tfdA sequence in this
class differed by 23% from class I. (Note that this percentage is less than the
25% difference when the entire tfdA sequence is taken into account.) Class III
tfdA sequences, which included strains TFD31 and 86-9, differed by 22.4%
from class I and 6.4% from class II. The alignment in Figure 1 includes the
tfdA sequences from IMP134, TFD39, and TFD23 as representatives of class I,
RASC as representative of class II and B6-9 as representative of class III.

It is interesting that both class II and III tfdA genes differed from each
other and from class I in the same nine sites, indicating probable
hypervariable hot spots in the wobble position.

Strain K1443, an alpha Proteobacterium which did not hybridize to
tfdA from JMP134 nor from strain RASC, (Ka, et al., 1994; and chapter 3)
yielded a 320 bp PCR band, although its intensity was only half that of the
other strains. However, it is not known whether this amplified DNA from
K1443 was from a functional tfdA analog, since it could not be aligned to the
other tfdA genes using Squd or other sequence alignment software,
including Sequencher from Genecodes and BLASTN and BLASTX (Altschul,
et al., 1990) from NCBI. Because the amplified fragment from K1443 did not
hybridize to tfdA from pIP4 even at low stringency, the tfdA nucleotide

125

[ 10 20 30 40 50]

I . . . . .]

M1 3 4 CATCACCAGATCICCCACG’I'I’CCAGCGA’IGCCGGTACACGAATTCCCGCT [ 5 O ]
'I'FD39 .................................................. [50]
TFDZB .................................................. [50]
869 ...... G ........ G T ....... T ...... A ..... A. .C T T [50]
RASC ...... A ..... C G T ....... T ...... A. G A C T G T [50]
[ 60 7O 80 90 100]

l . . . .]

JMPl34 GTGTCGCG'IGCI‘CGAGAAGCICCGCAAGCAGCATCCGGCCI‘ICGGCCACC [100]
TFD39 .................................................. [100]
TFD23 .................................................. [100]
869 C .................... A..C ........ G..A ........ G..G [100]
RASC C ..... A .............. T ........... G A ........ G. .G [100]
[ 110 120 130 140 150]

I . . . .]

M134 GGAAGGCC’I'I‘CGACG’IGGCI‘CGCG’IGCGCGCCGA’IGAAGAGAAACI'ICCG [150]
TFDB9 .................................................. [150]
TFDZ3 ................... C .............................. [150]
869 C . .C. .T...A.C A..A..C ..... A. .A ........... [150]
RASC C ......... T.. A A A..A..C ..... A. .A ........... [150]
[ 160 170 180 190 200]

l . . . . .]

M134 CCCGGAGCOGGOGI'CXQG‘I'ICGAACCAGCIBGCCAGTIGACCGGCCGCAW [200]
TFD39 .............................. A ................... [200]
'I'FD23 .............................. A ................... [200]
869 T. .A. T C ........ G. .G ..... T ..... AC.C ..... A ..... C [200]
RASC T. .A. T C ........ G. .G.T .......... C C ..... A ..... C [200]
[ 210 220 230 240 250]

[ I I O Q]

M134 CA'I'IGCG'I'IGCGCI'ICCGAATAG’IWI’CGCCGAGCAGGAAGCGBAG [250]
TFD39 .................................................. [250]
TFDZ3 .................................................. [250]
869 .......... ACT. . .A ........ C. .A ..... A .A.T ......... A [250]
RASC .......... ACI‘. ..A ..... C T ........ A..A.TA ........ A [250]
[ 260 270 280 290 300]

l . . . . .]

M134 WM [300]
TFD39 .................................................. [300]
TFDZ3 ................................... T .............. [300]
869 ..... C ........... T .G....AT..C.G..GT..C’I'I'C.T..AA.. [300]
RASC G...C ........... T ....... AT..C.G..G'1‘. CI'I’CTAA [300]
l 310 ]

l . ]

M134 CW [313]

TFD39 ............. [313]

TFDZ3 ............. [313]

869 TCG. .G. AT [313]

RASC 'I‘CG. .A. AT [313]

Figure 1. Alignment of 313 nucleotides of internal fragments of tfdA
genes from representative strains. Nucleotides identical to tfdA from
pIP4 are represented by periods.

126

sequence in K1443, if present, is probably less than 55% similar to that of pJP4.

A distance matrix was prepared from this alignment using 313 aligned
nucleotides. Only IMP134, TFD39, TFD23, RASC and TFD31 were used to
represent the other identical strains (Table 2).

A dendrogram of the tfdA genes was prepared from the aligned
sequences using maximum parsimony with MacClade (Figure 2). This shows
that the tfdA genes of IMP134, TFD39, TFD23, TFD9, K712, EML159 and TFD41
are all nearly identical. The class II and class HI tfdA genes are more closely
related to each other than either is to the class I tfdA gene. Because there is no
outgroup to the tfdA sequences, a rooted phylogeny could not be made, since
it not known whether the ancestral tfdA gene more closely resembles that
from IMP134, RASC or TFD31.

The different classes of tfdA gene were plotted on a map of North
America to show the geographic distribution of these genes. It is interesting
to note that identical tfdA sequences were found in organisms geographically
remote from each other (Figure 3).

Phylogenies are incongruent: The 16S sequence data described in chapter 2
were condensed to construct a phylogenetic tree using only those strains for
which tfdA sequence data is presented above. The most parsimonious tree
constructed using PAUP contained strain K1443 as an outgroup. This tree
has 13 strains with 10 unique 16S rDNA sequences, each differing by more
than 2 - 5%, indicating a high diversity of 16S rDNA for these strains. This
tree was then compared to the tree derived from tfdA sequence data (Figure
4).

A cursory observation of the most parsimonious phylogenetic trees for

Table 2. Distance matrix of tfdA-like genes from strains JMP134, TFD39,

TFD23, RASC and TFD31. Numbers above diagonal represent percent

similarities; numbers below diagonal represent number of base pair

differences in 313 nucleotides.

 

 

Strain JMP134 TFD39 TFD23 RASC 86-9
0413134 -- 99.7% 99.0% 77.0% 77.6%
TFD39 1 - 99.4% 76.7% 77.3%
TFD23 3 2 - 76.7% 80.0%
RASC 72 73 73 - 93.6%
B6-9 70 71 69 20 -

 

128

 

 

 

 

 

K712
RASC
TFD6
TFD2
E2W1
‘ 1.18
1
fl B69
TFD31

 

Figure 2. Dendrogram of 2,4-D degrading strains based on tfd A
sequences. Numbers on branches indicate numbers of
unambiguous base pair changes.

129

.vaECO $5 95 $me 3365 235 min «Em .Um<m ox: <35.“ 33:05 wasp—GENE
:wmum “oiwm 9E <35 firs mEmbm 23:25. 3.95%.. gm Smog—0L £28m Eot fl :2 «Es 3.95.5. 60.: mm 353: .3262): Bob
<3: 9 5:516. 6.8m :2: 32 32.238 <35 5:3 9:37. 33:65 mlwcflofl 22>» .mwcww <33 10 cousambflu uEmSwooU .m mhswE

 

1 (Bl)

.oocoscom <35 “0 mg Baa

WE who 00
.Ewfl of so E m3 how mac 9% :2 65 :0 fl <3 4‘ How Enuwepcmp < 5.8853

Evan op

Dim 3.826 Eofi <75“ m2 “Em wocow <3 A How Camawcgfi unocoworfm .v muswfi

r _._

mNuAHn~hr
..2:
..-6.M_

oDn—H

\Sonfi

3E2.
ones

m3

om~4_cm

 

3M.C,._._,

:2:
3547;
37:1

131

16S rDNA and for tfdA shows that the two trees are clearly different. For one
thing, the unrooted tree for tfdA shows only three, deeply divided branches,
representing three tfdA classes and five unique sequences, whereas the
phylogeny for 168 for the same strains show 10 unique sequences with fairly
deep branches, representing sequence differences between 2 and 18 percent.
Another way to show the relationship of tfdA genes to 16S rDNA was
to observe the distribution of tfdA classes among the branches of the 16S
rDNA phylogenetic tree. Figure 5 clearly shows that identical tfdA genes are
distributed throughout the beta and gamma Proteobacteria, and that adjacent
branches don't necessarily contain identical genes, as would be expected if no
interspecies gene transfer had occurred. Figure 5 shows several distantly
related strains (as determined by 168 rDNA comparisons) contain identical
tfdA genes. For example, strain RASC and TFD2 contain identical tfdA
sequences, yet their 168 sequences differ by 7%. A more dramatic
incongruency is seen in the case of TFD9 and TFD23. Although their tfdA
genes are identical, these strains differ in 45 of 283 165 rDNA nucleotides, or
16%. The degree of divergence in 165 genes is even more striking'when one
considers that the highly conserved 168 genes probably evolve at a much
slower rate (Moran et al., 1993; Wilson et al., 1977) than the tfdA genes.
Sometimes gene duplication within strains is a problem in
determining phylogenies (Doolittle and Brown, 1994) though this was not a
concern in this study. For example, it could be argued that two or more types
of tfdA genes could be found in individual strains, thereby confounding the
argument for gene transfer. However, this was not the case as supported by

two lines of evidence. First, PCR amplifications were made on total genomic

 

Burkholderia cepacia DS

1 32 Burldwlden'a cepacia G4

117039 0

 

Burkholderia mallei
Burklwlderia gladioli

   

Burkholder-1h mryophylli
Burkholderia andropogonis
‘

d RASC
B *7” .
IMPI34 .
TFDS8
‘ d ' Burldioldm'n solamceamm
. fimligmes eutroplms
Alailigenes x losoxidmxs
TFD9

 

 

 
  
 
  

r1323 0

ITEM I

869 I

Rhodofemxfermentans
" testoslosttroni

 

 

y Halomouas elougn la
Del. lmlophila
l-18

 

 

E.coll

   
 
 
 
 

24D1

Spliingomonns adlmesi'm
Sphingomoms paucinwbilus
Fltwobricl‘erium capsulatmn

3610
a K1443

EML146
Spliingomoms ymiolkuyae

  

 

Rhodopscudomorms pnlustris
M1

_ Desulfomonile tiedjii

Figure 5. Comparison of tfdA genes to phylogeny of 165 rDNA from 2,4-D degrading bacteria,
indicated in bold letters, and reference strains, indicated in italics. Colored lines indicate
phylogenetic groups of Proleobactmh: blue are beta, green are gamma, red are alpha, and black is
delta Proteobncten'n. Types of tfdA genes are indicated as either blue dots, green arrows, or red

133

DNA, and if there were duplicate, differing tfdA genes, sequencing would
have revealed ambiguous data, which was not the case. Secondly, only one
type of tfdA gene was observed for each strain in hybridizations studies
(discussed in chapter 3).
In vitro transfer of 2,4-D genes: To determine whether these 2,4-D degrading
strains could transfer the ability to degrade 2,4-D in the lab, mating
experiments were performed using B. cepacia D5 as a recipient strain. In
addition to the strains used for the sequence comparisons, other related
strains were added to test for their ability to transfer 2,4-D genes. These
included strains that hybridized to plasmid pIP4 and those that did not
(Matheson, 1994; Tonso, et al., 1995). To determine whether putative
transgenic strains could completely mineralize 2,4-D, they were grown in 500
ppm 2,4-D medium and 2,4-D disappearance was tested by HPLC. All
transgenic strains completely degraded 2,4-D as indicated by its disappearance.
Only some of the strains transferred and expressed the 2,4»D phenotype.
Interestingly, only those strains which had class I tfdA genes and which also
hybridized to plasmid pJP4 could transfer the 2,4-D phenotype to
Burkholderia cepacia D5. Those strains that belonged to class H or class III or
did not hybridize to pIP4 neither transferred nor expressed the 2,4-D
phenotype to B. cepacia D5 (Table 3), with the exception of strain RASC,
discussed below. The inability of the class II and class III tfdA strains to
transfer and express the 2,4-D genes may have been a result of the particular
conditions used here, such as an unsuitable recipient strain. However,
extensive mating experiments were conducted with one of these class II

strains, RASC (described in next chapter), which failed to transfer the 2,4-D

134

Table 3. Strains used for mating studies and results of in vitro 2,4-D phenotype
transfer.

 

 

 

PJP4
Strain Location Reference Hyb" Transfer@
IMP134 Australia Don and Pemberton, 1981 + +
EML159 (19) Oregon Amy et al, 1985 + +
TFD39 Saskatchewan Tonso et al., 1995 + +
K712 Michigan Ka, et al., 1994 + +
TFD9 Saskatchewan Tonso et al., 1995 + +
TFD41 Michigan Tonso et al., 1995 + +
TFD23 Michigan Tonso et al., 1995 + +
TFD31 Saskatchewan Tonso et al., 1995 - -
B6-9 Ontario Fulthorpe et al., 1995 nd nd
RASC Oregon Penny Amy - +
TFD6 Michigan Tonso et al., 1995 - -
TFD2 Michigan Tonso et al., 1995 - -
K1443 Michigan Ka et al., 1994 - -
EML146 Oregon Amy et al, 1985 -
2,4-D1 Washington Rob Sanford - -
TFD38 Michigan Tonso et al., 1995 + +
TFDZO Michigan Tonso et al., 1995 - -
TFD33 Michigan Tonso et al., 1995 + +
24L Michigan Tonso et al., 1995 - -
TFD34 Michigan Tonso et al., 1995 -
TFD27 Michigan Tonso et al., 1995 - -
'I'FD21 Michigan Tonso et al., 1995 - -
TFD13 Michigan Tonso et al., 1995 - -
'I'FDS Michigan Tonso et al., 1995 - -
TFD8 Michigan Tonso et al., 1995 - -
TFD4 Michigan Tonso et al., 1995 - -
TFD22 Michigan Tonso et al., 1995 - -
TFD14 Michigan Tonso et al., 1995 - -
TFD15 Michigan Tonso et al., 1995 - -
TFD12 Michigan Tonso et al., 1995 - -
26D Saskatchewan Tonso et al., 1995 + +
TFD18 Saskatchewan Tonso et al., 1995 - -
TFD29 Saskatchewan Tonso et al., 1995 + +
TFD42 Saskatchewan Tonso et al., 1995 + +
EML148 Oregon Amy et al, 1985 - -
* Strains whichhybridized to pIP4.

@ Strains which could transfer and express the 2,4-D phenotype in B. cepacia D5.

135

phenotype to a 2,4-D- sibling, suggesting that host incompatibility was not
necessarily responsible for limiting gene exchange.
Transfer of DNA from the 2,4-D degraders to P. cepacia D5 was

confirmed by hybridizing colony blots of donor, recipients, and at least three
different transgenic strains with the entire pJP4 plasmid labeled with 32R

Plasmid pIP4 was used as a probe for tracking gene transfer because it was
previously shown that these particular strains hybridized to pIP4 (Tonso, et
al., 1994). In all cases of transfer except one, the transgenic strains also
hybridized to plasmid pJP4 (Table 3). Figure 6 shows colony blots for a select
group of donor, recipient, and transgenic colonies probed with plasmid pJP4.
To confirm that the transgenic strains retained the recipient's chromosome,
colonies were also hybridized with the nptII internal gene fragment from
transposon Tn5 (data not shown).

One strain showed an unusual pattern of transfer. The only strain that
did not hybridize to pIP4 yet transferred its 2,4-D genes in vitro was strain
RASC. This strain was unusual because it transferred the 2,4-D genes only
very infrequently and at a very slow rate. Because it was so unusual, this
gene transfer event was studied in more detail as discussed in the next

chapter.

136

    

I!)
0 v- (\l to
.m .9 .9 .9
5 C C C
m a) a) a)
Q h on c» c»
‘1’ ° 2 2 2
°. 8 e 9 <2 .
m D l— l— +-

JMP134 "

K712 "

TFD41

TFD9

TFD38

TFD29

TFD33

TFD42

Figure 6. Colonies of B. cepacia DS, donors, and three
transconjugants were hybridized with total plasmid pIP4. This
demonstrates that a plasmid like pJP4 transferred in these strains.

137

DISCUSSION

This work has shown that gene exchange is responsible for generating
much of the diversity of 2,4-D degrading bacteria. The evidence for this is that
the tfdA genes have been transferred to and expressed in different species of
bacteria in nature, as indicated by the incongruent phylogenies of the tfdA
and 16S rDNA sequences for these strains. Evidence of interspecies gene
transfer can also be observed in that identical tfdA genes are found in strains
with 16S rDNA sequences differing by up to 16%, even though 16S rDNA is
regarded as a highly conserved gene and tfdA is unlikely to be conserved.

Further evidence for interspecies gene transfer is given by the results
that different tfdA genes were found in strains with the same 168 genes. For
example, strains TFD6 and EML159 contained identical 16S partial sequences,
yet their tfdA genes differed by 23%. Strains TFD23 and TFD31 also
contained identical 16S partial sequences, yet they also carried tfdA sequences
differing by 20%. Perhaps certain strains contain metabolic machinery, such
as a maleylacetate reductase genes (Kukor et al., 1989; Kaschabek and Reineke,
1992) which make them receptive hosts to a variety of different tfdA
sequences.

Barrier to gene flow or expression

The results of this study suggest a possible barrier to gene flow or
expression between members of the major subgroups of Proteobacteria.
Strains with tfdA sequences 55% or more similar to tfdA from pJP4 did not
appear to transfer from the beta group to the alpha subgroup, as indicated by
lack of gene transfer in laboratory experiments and the lack of hybridization

138

of any of the alpha group to tfdA(pIP4) (chapter 3).

This apparent barrier of gene flow between the beta and gamma and
alpha Proteobacteria is also found in plasmid pJP4, since no pJP4-like tfdA
genes were found among the alpha or gamma Proteobacteria. This is curious
since plasmid pIP4 is a broad host range plasmid and has been known to
transfer to alpha Proteobacteria, such as Rhizobium and Agrobacterium, and
to gamma bacteria such as Pseudomonas putida, P. fluorescens, and P.
aeruginosa (Don and Pemberton, 1981). However, in those laboratory
experiments, plasmids were transferred, but the 2,4-D pathways were not
expressed. This inability to express the phenotype may be due to lack of
accessory genes such as a maleylacetate reductase gene, required for complete
degradation of 2,4-D molecule (Kukor, et al., 1989) or to possible lack of 2,4-D
permease genes (Tim Sassanella, personal communication). The data from
nature, however, suggests that the transfer and retention of 2,4-D genes
between members of the alpha and beta subgroups of Proteobacteria probably
occurs rarely, since hybridization studies failed to show evidence of these
genes in the alpha Proteobacteria and since none of the members of the alpha
Proteobacteria hybridized to pIP4 (chapter 3 and Matheson, 1994, ).

Limited diversity of tfdA sequences

This study has shown only three basic types of tfdA sequences found
in strains belonging to 10 distinct 16S rDNA classes. Furthermore, Wright, et
al. (in prep) have sequenced two additional unique strains with tfdA
sequences identical to those of B6-9 reported here, including one from a
French agricultural isolate and another from an unculturable 2,4-D degrading

soil community. With the exceptions of the minor differences in various

139

pJP4-like tfdA sequences, why were no intermediate forms of these tfdA
sequences detected in these many unique strains? Presumably, these three
genes diverged long ago and only recently (perhaps only after massive
application of 2,4-D in the past half century) have these surviving tfdA genes
been particularly successful and have propagated throughout many
populations via gene exchange.

Broad host range transfer via plasmid and chromosome

In addition to demonstrating that gene exchange occurs in nature,
many 2,4-D degrading strains demonstrated gene transfer in the laboratory,
and the information from these in vitro mating experiments gave some clues
as to the mechanism of gene transfer. This study showed that in many cases
gene exchange was probably due to transfer of conjugal 2,4-D plasmids,
although another mechanism of gene transfer is suspected to have occurred
as well.

This study indicates that the class I tfdA genes (like those from pJP4)
are probably plasmid encoded. The evidence for this is threefold. First, all
donor and transconjugant strains hybridized to plasmid pIPI, whereas the
recipient strain alone did not. Second, all strains with a class I tfdA gene
examined so far contain broad host range, self transmissible plasmids
containing 2,4-D genes. For example tfdA genes are located on plasmids
found in JMP134 (Don and Pemberton, 1981), in TFD41 (Nakatsu, personal
communication), in EML159 (Amy, et al., 1985) and K712 (Ka and Tiedje,
1994). Third, all of the strains with a class I tfdA gene transferred the 2,4-D
phenotype at relatively high frequencies, a behavior more typical of conjugal

transfer rather than transformation of transduction.

140

The data in this study show that class I tfdA genes have a broad host
range among the beta Proteobacteria. These nearly identical tfdA genes have
been found in strains with 16S rDNA sequences differing from 7.6% up to
19% indicating a broad range of interspecies and intergeneric gene transfer.
Further evidence of the broad hOSt range of these strains is provided by the
fact that plasmid pIP4 and K712 are both members of the IncP plasmid
incompatibility group (data not shown, and Top, personal communication)
which includes highly self-mobilizable, broad host range plasmids (Thomas
and Smith, 1987).

Another type of tfdA gene which could be transferred in nature and to
a limited extent in laboratory experiments is that of strains RASC, TFD2, and
TFD6. The tfdA genes of these strains are probably chromosomally located as
appears to be true for RASC (Suwa et. al., 1994), TFD6 (Matheson et al., 1994)
and strain BRI6001 (Greer et al., 1991). The identical tfdA sequence is found in
strains RASC, TFD2, and TFD6 which differ in 16S rDNA nucleotide
sequences by 5.2 to 6.9 percent. Interestingly, this particular tfdA sequence
was found only in strains clustering into the genus Burkholderia,
determined by 16S rDNA sequences, but were not found outside this group,

' as were the class I tfdA genes.

The limited host range, among genus Burkholderia, of the class II tfdA
gene may be due to its limited ability to transfer and express 2,4-D genes, as
observed in laboratory experiments. The experiments performed here did not
eliminate the possibility that gene transfer was not observed due to the host

used. However, an unsuitable recipient may not have limited transfer since

141

strain RASC failed to transfer its 2,4-D phenotype to its sibling that was 2,4-D‘

(lacking a functional tde gene). None of the strains with a class II or class III
tfdA genes transferred and expressed the 2,4-D phenotype in lab experiments,
with the exception of strain RASC which did so at extremely low frequency.
Perhaps there is a barrier to gene transfer or expression for this type of tfdA
gene or its operon. This barrier may simply be due to the fact that this type of
tfdA gene is confined to the chromosome, or to an extremely large, relatively
immobile or integrated plasmid.

The third class of tfdA genes, which includes B6-9, I-18 and TFD31, may
have a very broad host range. Strain, I-18, a halophile isolated from a salt flat,
contains a tfdA gene identical to TFD31 and B6-9. Strain I-18 is a member of
the gamma Proteobacteria, whereas TFD31 and B6-9 are members of the beta
Proteobacteria. Despite the broad host range observed in natural isolates,
transfer was not duplicated in laboratory experiments Again, the lack of
observed transfer may have been due to incompatibility of a particular
recipient. It is not known whether TFD31 carries its 2,4-D genes on a plasmid,
however it is known that TFD31 carries no plasmid related to pJP4, as
indicated by its inability to hybridize to this plasmid (T onso, et al., 1994).
Anastomosis

Although this study includes a dendrogram based on partial sequence
from tfdA genes, this does not necessarily reflect the true phylogeny, that is
historical evolution, of these strains; rather, it may only reflect the phylogeny
of the tfdA molecules contained within these strains. Similarly, the
phylogeny of the 168 genes may not necessarily be construed as the true

142

phylogeny of these organisms, since organisms are amalgams of different
genes. Instead, the tfdA and 16S rDNA sequence data indicate that the
evolution of these strains has been by a process of "anastomosis", (or
reticulate evolution) in which branches of the phylogenetic tree merge,
forming a network (Margulis, 1992).

The 16S rRNA molecule has revolutionized our understanding of
prokaryotic phylogeny, but until at least some portion of the remaining 99.99
percent of genes in any strain are also analyzed for comparable phylogenies,
we should be cautious in relying solely on 168, which Olsen et al., have called
"a key to prokaryotic phylogeny" (1994), for determining the phylogeny of the
entire organism. Many microbiologists, including myself, seem to be swept
up by the "Winds of (Evolutionary) Change," (Olsen, et al., 1994). However,
in the enthusiasm for having discovered a possible Rosetta stone for
deciphering the evolution of all life, we should not overlook the fact that 16S
rRN A sequences account for only a small portion of the entire genome. In
the future, comparing the phylogenies of other genes to the phylogenies
derived from 16S rRNA should address this concern.

So far, phylogenies based on ribosomal sequence data have only been
corroborated by a handful of other molecules, including elongation factors EF-
1 alpha (Tu) and EF-2(G), and the alpha and beta subunit of ATP-synthase
(Iwabe, et al., 1989) and cytochromes (Woese, et al, 1980), and in a handful of
others in a limited number of strains (Doolittle and Brown, 1994). Even with
these data there still exists the problem of gene duplication and horizontal
gene transfer (Doolittle and Brown, 1994) and inability of current methods to
discern a true phylogeny (Doolittle and Brown, 1994, Creti, et al., 1994).

143

In addition to corroborating phylogenies created with ribosomal data,
researchers have performed phylogenetic congruency tests on other
molecules. Dykhuizen and Green found that the phylogenies of the phoA,
gnd, and trp genes were incongruent (Dykhuizen, 1991). However, their
study was confined to one species, where (by definition in the
eukaryotic / sexual sense) horizontal gene transfer can occur.

Some may object to using "accessory genes," such as genes for 2,4-D
degradation, for comparing phylogenies of organisms, because similar genes
are not necessary under all circumstances. Although accessory genes are not
essential to bacteria under all conditions, these genes are absolutely essential
for survival under some conditions and may comprise a significant portion
of the bacterial genome. For these reasons it is only reasonable to include the
ancestry of these accessory genes in the phylogeny of an organism. Though
this task is orders of magnitude more complex and less convenient for
systematists who have enough of a challenging task to determine the
phylogeny of a single gene, it is a more accurate representation of reality.
Phylogeny of the gene is not necessarily the phylogeny of the organism

Prior to systematics based on 16S sequencing, the primary means of
categorizing bacteria was phenotypic. In our enthusiasm for the simplicity of
a system based on 165, we should not forget that phenotypes are also based on
genetic data, which are as amenable to phylogenetic analysis as is ribosomal
DNA. To infer that strain A is more like strain B than like strain C based on
similarities of their 165 genes is like concluding that a Mitsubishi pickup is
more related to a Mitsubishi toaster than it is to a Ford pickup, just because

the former two share a Mitsubishi microchip for heater control.

144

A true phylogeny of an organism can only be inferred when the
phylogenies of several genes per organism have been compared. Comparing
ribosomal RNAs has created a revolution in understanding evolution and
comparing the phylogenies of more molecules will lead to an even bigger
revolution in understanding microbial evolution.

This study has shown that interspecies gene transfer is important in
the evolution of 2,4-D degrading bacteria. The evolution of these organisms
cannot be inferred by examining 16S rDNA alone or tfdA genes alone, but
when examined together, these two genes present a clearer picture of the

evolution of these bacteria in nature.

ACKNOWLEDGEMENTS

I thank Dr. William Holben for 2,4-D strains and tfdA, tde and tde
gene probes and Dr. Iong-Ok Ka, Dr. Nancy Tonso, Dr. Penny Amy and Dr.
Charles Greer for the use of their strains. I would like to thank Dr. Richard
Lenski for valuable discussions in comparing phylogenies of tfdA and 16S
rDN A and for encouraging me to partially sequence the tfdA genes. I
especially thank Dr. Alice Wright and Dr. Roberta Fulthorpe for their
assistance in sequencing the tfdA genes. This work was supported by the
National Science Foundation grant number BIR912006 and by the Japan

Research and Development Corporation.

Chapter Five

Interspecies Transfer of Chromosomally Encoded
2,4-Dichlorophenoxyacetic Acid-Degrading Genes

145

Interspecies Transfer of Chromosomally Encoded

2,4-Dichlorophenoxyactic Acid Degrading Genes
SUMMARY

Identical chromosomally located tfdA genes, which encode the first
enzyme for the degradation of 2,4-dichlorophenoxyacetate, have been found
in phylogenetically diverse strains, yet it is not known by what mechanism
these genes have transferred. Here it is shown that the tfdA and B genes
have were transferred from the chromosome of strain RASC to the
chromosome of Burkholderia cepacia strain D5 in the laboratory. Transfer of
the 2,4-D genes was not via the only detectable plasmid (11 kb) harbored by
RASC. Results from independent mating experiments indicated that the
genes inserted into different sites in the recipient B. cepacia D5 chromosome.
The transferred 2,4»D phenotype appeared stable indicated by the fact that
serial transfers of RASC and transgenic strain 8. cepacia D5(RASC) in
nonselective medium over 300 generations showed no loss of the 2,4-D
phenotype. Integration of the 2,4-D genes into the new host's chromosome
may have involved the concomitant transfer of a Repetitive Extragenomic

Palindromic (REP) fragment from RASC into the new host.

146

147

INTRODUCTION

It is thought that the broad diversity of chloroaromatic degrading
organisms has evolved via transfer of degradation genes (Fox,1983; Kellogg et
al., 1981; Yano, 1980). Generally, this transfer is thought to occurred via
conjugation (Sayler et al., 1990). Chakrabarty and Gunsalas first found that
camphor degradative traits were on plasmids, and since then, genes for
degrading many chlorinated aromatic compounds have been found on
plasmids. Until now, the only reported means for transfer of such genes to
other organisms is via conjugation of plasmids ( for review see Sayler, 1990),
or via transposons residing at least temporarily on plasmids (Wyndham, et
al, 1988; Fulthorpe, et al., 1991; Nakatsu, 1992; Burlage et al., 1989; Ieenes et al.,
1982). So far there has been no report of transfer of catabolic genes from the
chromosome of one strain into the chromosome of another.

Plasmid mediated transfer of 2,4-D degradation has been well
documented, beginning with the best characterized 2,4-D plasmid, pJP4 from
Alcaligenes eutrophus IMP134 (Fisher, et al., 1978; Don and Pemberton, 1981;
Friedrich, et al., 1983). Since the discovery of plasmid pIP1, several other
researchers have reported 2,4-D genes carried on plasmids in a broad range of
organisms (Don and Pemberton, 1981; Chaudry and Huang, 1988; Ausmees,
et al., 1990, Amy, et al., 1985; Bhat, et al., 1994; chapter 4) whereas there have
been only suggestions of chromosomally encoded 2,4-D genes ( Amy, 1985;
Greer, 1990; Ka, 1993; Suwa, 1994). Genes for 2,4-D degradation have been
shown to transfer only via conjugation of plasmids (Don and Pemberton,

1981; Amy, et al., 1985; Bhat, et al., 1994; Chaudry and Huang, 1988).

148

However, a broad range of 2,4-D degraders with apparently chromosomally
encoded 2,4—D genes may be more widely distributed than strains harboring
plasmid-borne, 2,4-D-degrading strains (Matheson, et al., 1994; Chapter 2).
The genes in these strains are 60—80 percent similar to those of pJP4,
(Fulthorpe, et al, submitted; Matheson, et al., submitted; Suwa, et a1. 1994) yet
they appear to be on the chromosome (Suwa, et al., 1994.)

In this study we investigated the ability of these apparently
chromosome-borne 2,4-D genes to be transferred. Here we present evidence
for dissemination of the 2,4-D phenotype from the chromosome of strain ‘
RASC to the chromosome of B. cepacia which suggests a paradigm for transfer

of 2,4-D genes in nature other than by conjugation.

MATERIALS AND METHODS

Media and growth conditions: 2,4-D MMO containing 500 ppm 2,4-D (Sigma)
in a mineral salts medium (Stanier, 1966 ) was used to grow 2,4-D degrading
strains. Antibiotics including kanamycin, carbenicillin, and bacitracin, were
added at concentrations of 50 ug/ ml. Peptone, tryptone, yeast extract, glucose
(PTYG) medium is a dilute nutrient broth containing 0.25 g peptone, 0.25 g
tryptone, 0.5 g yeast extract and 0.5 g glucose, 0.03 g magnesium sulfate, and
0.003 g calcium chloride per liter. TSB contains 15 g Trypticase soy broth
(Difco, Detroit, MI) per liter. All cultures were grown at 30°C. Broth cultures
were aerated by shaking on an Innova 4300 shaker (New Brunswick Scientific
Co., New Brunswick, NJ) at 200 rpm.

Bacterial strains. Strain RASC was isolated by Dr. Penny Amy from return

149

activated sludge from the Corvalis, Oregon waste water treatment plant (Dr.
Penny Amy, personal communication). It is sensitive to the antibiotics
kanamycin, carbenicillin, and bacitracin. B. cepacia strain D5 was supplied by
Dr. Ron Olsen (Walch and Ballou, 1983). It was made kanamycin resistant by
the introduction of neomycin phosphotransferase genes carried on Tn5
(Holben, personal communication) into a nonessential region of B. cepacia
strain DBOl. This strain is resistant to 50 pg / ml each of kanamycin,

bacitracin and carbenicillin. B. cepacia DBOl 13R1 is a spontaneous rifampicin
resistant, methionine auxotroph of B. cepacia D5. RASCd2 is a 2,4-D‘ mutant

made by Tn5 interruption of the tde-analogous gene of RASC (Suwa, 1994).
Escherichia coli (pR388) contains the Inc W plasmid pR388 which is resistant
to streptomycin. Alcaligenes eutrophus IMP134(pIP4) contains 2,4-D genes
located on its 80 kb plasmid (Don and Pemberton, 1981) and was provided by
Dr. Ron Olson.

Mating Studies. A ten-fold excess of recipient B. cepacia D5 in broth (~ 109

cells) was mixed with RASC donor cells and centrifuged briefly to concentrate.
These cells were then pipetted either onto 0.2 um filter paper, (Whatman),
which was placed onto agar plates, or onto agar media directly. Strain RASC
and B. cepacia D5 were mated under diverse conditions of varying surfaces
and substrates (agar, broth, or filters), duration, and media. The donor strain
was grown either in TWIG, TSB, or 2,4-D. The recipient strains were grown
either with or without kanamycin in TSB or PTYG. Agar plates contained
2,4-D with or without antibiotics, PTYG or TSB. Cells were allowed to
incubate together for various times ranging from 2 hours to 16 days. Mating

150
mixtures were then either resuspended in H20 and serial dilutions made

before plating, or were streaked directly onto selective media of 2,4-D
kanamycin, carbenicillin and bacitracin. IMP134 was mated with B. cepacia
D5 as a positive control. I

2,4-D Biodegradation. Disappearance of 2,4-D was monitored by HPLC.
Supematants were centrifuged at 14,000 rpm and filtered through 0.2 u filters.
Samples were then analyzed by HPLC on a Lichrosorb Rp-18 column (Anspec
Co., Ann Arbor) using 60% methanol and 40% 0.1% H3PO4 as eluant. 2,4-D

was detected at 230 nm.

Transformation studies. An excess (5 ug) of DNA from strain RASC was
added to 1 ml (~ 108 cells) of B. cepacia D5 concentrated to 100 pl by brief

centrifugation and incubated either in PTYG broth, half strength TSB plates
or on 2,4-D MMO with 50 ug/ ml each kanamycin, carbenicillin, and
bacitracin. After one week incubation, a loopful from each medium was
streaked onto agar plates containing 2,4-D MMO with and without 50ug/ ml
each of kanamycin, carbenicillin, and bacitracin. Putative transformant
colonies were tested purified by streaking onto fresh 2,4-D plates.

DNA preparation and restriction analyses. DNA was isolated using the
procedure of Ausubel, et al., 1992. DNA was digested according to
manufacturers directions using restriction enzymes from Boehringer
Mannheim. These included EcoRI, Natl, HindIII, SalI, PstI, PvuII, BglII, SauII,
Smal, SfiI, and SwaI. DNA molecular weight markers included
bacteriophage lambda digested with HindIII (Boehringer Mannheim),
MegaBase II marker (Biorad, Hercules, CA), and 100 bp marker (Bethesda
Research Laboratories, Gaithersburg, MD). DNA was transferred to nylon

1 51
membrane (Amersham Hybond-NTM) and covalently linked to the

membrane with ultraviolet radiation using a StratalinkerTM 1800

(Stratagene, La Jolla, CA).

Hybridization. The 11 kb plasmid, pRASC and the nptII gene were labeled
using random hexanucleotides incorporating (32P)dCTT’ (3,000 Ci / mmol; New
England Nuclear, Boston, MA). DNA probes were made from internal
fragments of the tfdA, B, C, and D genes from plasmid pIP4 (Holben, et al,
1993) and from the 3.0 kb and 2.2 kb fragments from strain RASC which
included the tfdA-like gene from this strain (Suwa, in press). Plasmids
containing DNA fragments were digested with appropriate restriction
enzymes, purified by gel electrophoresis, excised from agarose, and labeled
with digoxygenin-l 1-dUTP using the Random Primers Labeling Kit
(Boehringer Mannheim) according to the manufacturer's instructions.
Hybridizations using probes from IMP134 were carried out under low
stringency conditions using no formamide and a temperature of 62°C.
Hybridization with the probes from RASC were carried out under high
stringency conditions using 50% formamide and 62°C as described in chapter
3.

Phenotype stability. Stability of the 2,4-D phenotype was examined by
grong RASC and D5(RASC) for 300 generations by transferring 25 ul
(approximately 1.7 x 106 cells as determined by plate counts) into 25 ml non-
selective broth (PTYG) each day for 30 days. Strains IMP134 and B. cepacia
D5(pIP4) were treated in the same manner as a control. After each 24 hour
period cells were counted by plating, and found to number approximately
1.7 x 109 cells per m1.

152

The number of generations per day was calculated as :

(log Populationfinal - log Population initial) / log 2
The number of generations per day, found to be 10, was multiplied by 30 days
resulting in 300 generations total. Colonies were counted on non-selective
PTYG medium and on 2,4-D medium on days 1, 2, 7, 8, 9 and 30.
Large Plasmid Detection. Protocols of Kado and Liu, (1981) of Wheatcroft et
al., (1990) and of Eckhardt (1978) were used and modified for trying to detect
large plasmids from strain RASC using IMP134 as a positive control. The
largest plasmid yield for pJP4 was obtained using the modified method of
Kado and Liu. Overnight culture grown in TSB was pipetted into a 2 ml
microfuge tube and pelleted by centrifugation at 14,000 rpm. Cells were
resuspended in 200 pl of Buffer E (Kado and Liu, 1981) and mixed with 400 pl
of lysis buffer. The lysate solution was incubated at 68°C for 1 hour. A
volume of 600 1.11 of phenolzchloroform (1:1) was added and the lysate was
inverted to mix. The lysate was then centrifuged at 14,000 in 4°C and the
supernatant was transferred to a clean microfuge tube. Plasmids were
precipitated by addition of 100 pl of 3 M NaAcetate, pH = 5.3, and 700 pl
isopropanol and then stored overnight at -20. DNA was pelleted by
centrifugation and pellets were resuspended in 40 ul T10E1 and then

separated by gel electrophoresis on a 0.7% agarose gel.

REP-PCR - DNA from strains was subjected to PCR amplification using
reverse extragenomic palindromic (REP) primers (Right =
5'GCCIGATGICGICGI-3' and Left = 5'-ICGICTTATCIGGCCTAC-3')
synthesized by the Macromolecular Sequencing Facility at Michigan State
University (deBruijn, 1993). Reagents for PCR were added according to the

153

protocol of deBruijn, 1993. Taq polymerase was purchased from Perkin
Elmer. Thermal cycling was done in a Perkin Elmer 9700 Thermal Cycler

using the following conditions: Melting at 95°C 1 min; cycling 35 times at
940C for 1 min, 40°C for 1 min, and 65°C for 2 min followed by an elongation
time of 8 min at 65°C. PCR products were separated by electrophoresis on a
1.5% agarose gel (Boehringer Mannheim) at 38V for 12 hours using a 100 base

pair ladder as a size marker. The gel was stained with ethidium bromide and

photographed with Polaroid positive negative film 55.

154

RESULTS

Mating experiments. Strain RASC and B. cepacia were initially mated on a
cellulose filter overlying nonselective TSB agar, incubated 25 hours and then
serially diluted and spread plated onto selective media containing 2,4-D as the
sole carbon and energy source and 50 ppm each of carbenicillin, bacitracin,
and kanamycin to select against the donor. Negative controls with donor
cells or recipient cells alone yielded no colonies on the selective medium.
Colonies of donor cells from mating mixture were counted on 2,4-D medium
and numbered 2.6 x 108 donors per ml. Colonies of recipient cells in mating
mixture, counted on PTYG 50 ppm each of carbenicillin, bacitracin, and
kanamycin, numbered 2.1 x 109. Transgenic strains were observed only after
four weeks incubation on selective plates. The frequency of transfer was very
low and could not be reliably quantitated since several plates with dilutions of
10-1, 10-1 and 10-2 contained approximately the same number of colonies (1, 66
and 2 colonies, respectively). This is a frequency of between 2.5 x 107 and 3.8 x
10-7 transgenic strains per donor. In contrast, a positive control mating with
EML159 and B. cepacia D5 yielded 2 x 10-4 transconjugants per donor which
appeared after only three days. In another control experiment, IMP134 mated
with B. cepacia D5 yielded 50% transconjugants per donor.

Transgenic colonies were transferred onto fresh selective agar three
times to isolate pure colonies. The colonies grew more quickly with each
successive transfer.

To determine whether the putative transgenic strains could completely

155

mineralize 2,4-D, they were grown in 500 ppm 2,4-D medium and 2,4—D
disappearance was tested by HPLC. All transgenic strains showed 100%
disappearance of 2,4-D and no appearance of any other U.V. absorbing peaks.
Transfer of DNA. Transfer of DNA from strain RASC to B. cepacia D5 was
confirmed by hybridizing Southern blots of EcoRI digested donor, recipient
and transgenic strains with the tfdA probe from pJP4 at low stringency. A 7 kb
band appeared on lanes with both RASC and D5(RASC), but not on the lane
of the recipient strain B. cepacia D5. Hybridization with tde yielded the same
pattern with the appearance of an additional 3.5 kb band in both donor and
transgenic strains (Figure 1). Interestingly, the tde and tde gene probes
showed no hybridization signals, even at very low stringency. A probe
constructed of a 3 kb fragment from RASC containing the tfdA-like gene from
this strain also hybridized to the 7 kb EcoRI fragment (Figure 2).

To determine whether transfer of the 2,4-D phenotype was
concomitant with the transfer of the resident 11 kb plasmid in RASC, the 11

kb plasmid from RASC was labeled with 32P and used to probe donor,

recipient and transgenic strains (Figure 3). Since the plasmid did not
hybridize to the transgenic strain, this indicated that the genes were not
located on the 11 kb plasmid. The conclusion was subsequently confirmed by
Yuichi Suwa in this lab by plasmid curing experiments.

To see if the transferred genes were inserted in the chromosome, total
genomic DNA from donor and transgenic strains was digested with a seven
different restriction enzymes and probed with a 3.0 kb fragment of DNA from
RASC that contained both tfdA- and tde-like genes. Two of the restriction
enzymes, 83111 and PvuII yielded different restriction patterns, indicating that

156

DS
JMP134

o
a)
<
5
LO
0

 

Figure 1. Southern blot of EcoRI-digested total genomic DNA from RASC, B.
cepacia D5, and B. cepaciu D5(RASC) was hybridized at low stringency with tde
from pJP4. This probe hybridized to the same 7.0 kb and 3.5 kb fragments in
RASC and D5(RASC), indicating transfer of genetic material from one strain to
the other. This hybridized with an intense signal to the positive control JMP134,
but not to the negative control B. cepacia D5.

157

B. cepacia D5

RASC
D5(RASC)1

V1
('0
g

 

Figure 2. DNA from jMP134, RASC, D5, and D5(RASC) were
digested with EcoRI and hybridized with the 3.0 kb fragment from
RASC which contains the tfdA-like gene.

158

 

 

 

 

 

 

 

 

 

Figure 3A. Duplicate colonies of RASC, B. cepacia D5 and
transgeneic strains B. cepacia D5(RASC) were hybridized
to the 11 kb plasmid isolated from strain RASC. The blot
in figure B was hybridized to nptII from Tn5, which is
contained on B. cepacia D5. The labeled plasmid from
RASC did not hybridize to the transgenic strain indicating
that the genes for 2,4-D degradation were not located on
this plasmid.

159

the 2,4-D genes had either inserted into a site in the recipient's chromosome
or rearranged after entering the new host (Figure 4).

Stability of 2,4-D phenotype. After more than 300 generations in
nonselective broth medium 100% of the strains of RASC, D5(RASC), and
IMP134 or D5(pIP4) (used as controls) retained the 2,4-D degradation
phenotype. This stability indicates that the genes are not on an unstable
transposon or plasmid, but are located on the chromosome, or on a very
stable transposon or plasmid.
' Reproducibility. To reproduce and optimize this gene transfer event, 29
different mating conditions were designed to optimize the frequency of DNA
transfer from strain RASC to B. cepacia. These different experiments
employed different media, different incubation times and different surfaces to
optimize conditions for transfer (Table 1). Of these 29 independent mating
experiments, transgenic strains were observed in only six, and these only after
a period of incubation on plates for at least three weeks. Media type and
mating support (i.e. filter or agar) appeared to be less important to gene
transfer than sufficient time.

To determine whether the 2,4-D genes could be transferred back into
their original host, transgenic strain D5(RASC) was mated with a 2,4-D Tn5

mutant of RASC unable to grow on 2,4-D. This produced no transgenic
strains capable of growth on 2,4-D.

To see whether the genes could be transferred from B. cepacia
D5(RASC) to a sibling of B. cepacia D5 , transgenic strain D5(RASC) was mated
with B. cepacia DB0113R1, a spontaneous rifampicin resistant, methionine

auxotrophic mutant of DBOl. This produced no transgenic colonies.

160

 

Figure 4. Total genomic DNA from RASC and B. cepacia D5(RASC) was
digested with seven restriction enzymes and then probed with tfdA from
RASC. The restriction enzymes Pvull and BglII yielded different restriction
fragments, indicating that the DNA had a different genomic arrangement in
B. cepacia than in strain RASC.

161

Table 1. Conditions used in matings

 

 

Br M ia

RASC D5 Support Agar Time‘
PTYG PTYG Filter/ 24 h PTYG 3.5 weeks
PTYG PTY G Plate/ 24 h PTYG -
PTYG/2,4D PTYG Plate 2,4-D KCB -
PTYG/2,4D PTYG Replica Spot 2,4-D KCB 3 weeks
PTYG PTY G Spot 2,4-D KCB 3 weeks
PTYG PTYG Filter/ 24 PTYG -

TSB TSB Filter/ 2 weeks PTYG -

TSB TSB Filter/ 2 weeks PTYG -

TSB TSB Spot/ 16 days 2,4-D,HMK 2 weeks
2,4-D PTY G Filter/ 24 h 2,4-D -

2,4-D PTYG Filter/ 24 h PTYG -

2,4D PTYG K1 Filter/ 48 2,4-D KCB -

2,4D PTYG K Filter / 48 PTYG -

2,4-D PTYG K Filter / 48 2,4-D KCB -

2,4-D PTYG K Filter/ 48 TSB/2,4D 5 weeks
2,4aD PTYG K Filter/ 2 2,4-D -

TSB TSB K Filter/ 2 TSB/2,4D -

TSB TSB K Filter/ 24 TSB/ 2,4D -

TSB TSB K Filter / 48 TSB/ 2,4D -

TSB TSB K Filter/ 3 2,4D KCB -

TSB TSB K Filter/ 24 2,4D KCB 5 weeks
TSB TSB K Filter/ 48 2,4D KCB -

TSB TSB K Filter/ 4 2,4D -

TSB TSB K Filter / 24 2,4D -

TSB TSB K Filter/ 48 2,4D -

2,4-D TSB K Filter/ 3 T'SB/ 2,4D -

2,4-D TSB K Filter/ 24 TSB / 2,4D -

2,4-D TSB K Filter/ 3 2,4D -

2,4-D TSB K Filter/ 24 2,4D -

 

* Time at which transgenic colonies were first observed. A "-"

indicates no transgenic colonies were seen.
1) "K" indicates addition of kanamycin

162

To see whether the same restriction patterns were found in
independent transgenic strains, DNA from two mating experiments was
isolated and digested with BglII (Figure 5). The different restriction fragment
lengths of the two transgenic strains indicate that unique transfer events
yielded unique insertion sites.

Transposon Trapping. To investigate the possibility that the transferred genes
might be on a transposon, transposon trapping experiments were performed
using plasmid p388, a broad host range Inc W plasmid as the "trap"

(Valentine and Kado, 1989; Bennett and Grinsted, 1988). This plasmid,
carrying streptomycin resistance, was mated from a proline deficient
auxotrophic strain of E. coli into RASC. The mating mixture was plated on
MMO agar containing 2,4-D and streptomycin to recover transconjugants.
This transfer occurred at a frequency of 61% transconjugants per donor.

These resulting transconjugant RASC(pR388) strains were then mated with B.
cepacia D5. No 2,4-D degrading transconjugants were observed. The
experiment was repeated with the RASC(pR388) strains stressed by starvation
and transconjugants were still not observed.

Transformation studies. To determine whether DNA from strain RASC was
transferred via transformation, DNA isolated from strain RASC was added to
cultures of B. cepacia D5 and allowed to incubate for one week. In one
experiment the cells and DNA were left to incubate on selective agar plates;
the second incubation was on nonselective agar (one half strength TSB) and
in the third experiment cells and DNA were incubated in PTYG broth
without mixing. In three experiments no transformants were found.

Transfer of REP fragment. Strains of donors RASC, recipient D5, and seven

163

2.9.9.
UUU
U 3 33
mvvv
8 § 8 B B

—23.1

—9.4

—6.7

—4.4

_2.3

—2.0

 

Figure 5. Total genomic DNA from P. cepacia D5, RASC and 3
separate transgenic strains was digested with Bng and hybridized
with a 3 kb fragment from RASC containing its tfdA like gene as well
as a portion of Tn5. DNA from RASC hybridized to different size
fragments in the transgenic strains, indicating different insertion sites
in each as a result of different gene transfer events.

164

transgenic strains of D5(RASC) were PCR amplified using repetitive
extragenomic palindromic (REP) sequences as primers. Banding patterns for
B. cepacia D5 and D5(RASC) differed from that of strain RASC, as expected.
However, 8. cepacia D5 and D5(RASC) shared identical banding patterns
except that all the transgenic D5(RASC) strains shared an additional 1.2 kb
fragment, not found in B. cepacia, not found in strain RASC (Figure 6).
Hence, it appeared that a 1.2 kb fragment from RASC amplified by REP PCR
had transferred to B. cepacia D5.

To confirm that this apparent new band was from RASC, and not an
artifact of PCR, the 1.2 kb band was excised from the gel, labeled with
digoxygenin, and used to probe a Southern blot of the PCR amplified REP
fragments (Figure 7). This 1.2 kb probe hybridized only with RASC and the
transgenic strains of D5(RASC), and not to B. cepacia D5.

To investigate whether this PCR amplified REP fragment was directly
adjacent to the tfdA or tde genes in RASC, DNA from RASC, B. cepacia D5,
D5(RASC) and JMPl34 digested with EcoRI, separated by gel electrophoresis,
blotted onto nylon filters and probed with the 1.2 kb fragment from RASC
(Figure 8). The resulting fragment size which hybridized to this probe was 4.7
kb, much smaller than the 7.2 and 3.5 kb EcoRI fragments containing the
tfdA and tde-like genes.

165

1
2
3
4
6
5
7

DS RASC

.s
U
(U

QQQQ.%
U

1.2 kb

 

Figure 6. Strains RASC, B.cepacia D5 and seven transgenic
strains of B. cepacia D5(RASC) were PCR amplified using REP
primers. RASC contains amplified bands differing from other
strains. The transgenic strains all had the same size fragments as
those of B. cepacia D5 with the addition of a 1.2 kb fragment
shared with strain RASC.

166

FNMVCDIDN
8688888
UJUJUJCDCDCDCD
<<<§<<<
EEEVEEE

$3 mmmmmmmm
5 00000000
I gggggggg
6 33333333
3088888888
Egoooooooo
Sammmqmmgm

 

Figure 7. REP-PCR amplified DNA from strains RASC, B.
cepaciu D5 and 7 transgenic strains of B. cepacia D5(RASC) was
blotted and probed with a 1.2 kb fragment from strain RASC.
Only strain RASC and the transgenic strains hybridized to this
fragment.

167

 

    

E :5"
-- MM
i 3 m E E
B :2 9 H H U
E O CO V)
NEmMm<
_1 DOOM

13:3:

y—I

wg

«bi

L‘——.=-.

to“; ”as... _4'7kb

”E

Gaga

‘é'

y“:

* limit

6
ll

Figure 8. Hybridization of EcoRI digested total genomic DNA with a 1.2 kb
REP amplified fragment isolated from RASC which appears to have
transferred into separate transgenic strains of B. cepacia D5(RASC).

168

DISCUSSION

The tfdA and tde genes in the 2,4-D'degradation pathway transferred
from the chromosome of strain RASC into the chromosome of B. cepacz'a D5.
This was shown by the fact after RASC and D5 were mated, colonies appeared
on selective media containing 2,4-D, kanamycin, carbenicillin, and bacitracin,
whereas controls of either strain alone did not. Furthermore, these
transgenic strains were able to mineralize 2,4-D completely when cultured in
broth (unlike each of the parent strains alone). It was also shown that the
transgenic strains hybridized to the tfdA and 1:de probes from IMP134 and
from strain RASC, whereas the recipient strain did not.

The transfer and expression of the 2,4-D phenotype occurred
infrequently and very slowly in vitro. At least 15 kb of DNA transferred
from RASC into D5 as judged by the sizes of transferred restriction fragments.
The gene transfer could also be reproduced, and was in six different
experiments. It was also shown that different transgenic strains had tfd genes
inserted into different sites in the chromosome, as indicated by different RFLP
patterns.

It is not surprising that DNA from strain RASC transferred to and was
expressed in a strain of B. cepacia, since strain RASC is member of the genus
Burkholderia (Y abuuchi, et al., 1992) as determined by 16S rRNA (chapter 2).
Strain RASC has been shown to be very similar to another 2,4-D degrader
(isolated from a distant part of the US.) strain K712, differing by only 2% in
165 genes (Ka, et al., 1994; chapter 2). Interestingly, a close relative of strain

169

K712 has 2,4-D genes located on a plasmid which has been shown to integrate
into the chromosome, (Ka, et al., 1994d).

It is perhaps not surprising that 2,4-D genes could be expressed in B.
cepacia since other strains of B. cepacia have previously been shown to be
involved in 2,4-D degradation (Bhat, et al., 1994; Greer, et al., 1990) . B.
cepacia CSV9O has been discovered which contains a 2,4-D plasmid with a
tde gene identical with that of pJP4 (Bhat, 1994). Another B. cepacia strain,
BR16001, which was isolated from peat in Quebec, Canada, reportedly has 2,4-
D genes located on the chromosome. Interestingly, strain BRI6001 contains a
tfdA gene which is identical to that of the tfdA gene of strain RASC (chapter
4).

Mechanism of transfer. This story would be more satisfying if we knew the
mechanism of transfer or of integration of the 2,4-D genes. However, since
the gene exchange and expression required such a long incubation time, there
is only little data on gene exchange mechanisms. Still, some observations
were made and I would like to present them here.

The slow rate at which the 2,4-D genes transferred to B. cepacia may be
a clue to the mechanism of gene transfer or of integration into the
chromosome. The long incubation time required before the transfer and
expression of the 2,4-D genes could be the result gene rearrangement before
expression of the 2,4-D degradation phenotype. Perhaps the long incubation
time was required for the lysis of RASC and uptake of DNA into B. cepacia
D5. Another possibility is that the transfer event occurred quickly, but the
expression required time for appropriate gene rearrangement in the new host.

Many researchers have observed a long period of adaptation of microbial

170

communities before expression of a particular biodegradative phenotype
(reviewed by van der Meer, et al., 1992), which can be attributed to genetic
recombination. Because the restriction patterns of donor RASC and
transgenic B. cepacia D5 (RASC) differed, this suggested genetic recombination
in the new host. In addition, the slow time for expression of the 2,4-D
pathway in all six transfer events suggested the possibility that the genes
required time to rearrange themselves into a proper orientation before
properly expressing the 2,4-D phenotype.

Several experiments were designed to investigate the possibility of
conjugation as a mechanism for transfer, and all were negative. First, the
only readily detectable plasmid in strain RASC, an 11 kb plasmid, did not
transfer, although the 2,4-D phenotype did. Second, several different
protocols were used to detect large plasmids in strain RASC and the
transconjugant. Although plasmids from the positive control, JMP134, could
be detected, no large plasmids were detected in strain RASC. Another line of
evidence that may rule out the possibility of conjugation is the fact that the
2,4-D phenotype did not transfer from B. cepacia D5 back into strain RASC nor
into another strain of B. cepacia. Conjugation also seems unlikely due to the
long incubation times required (discussed above).

Another possible mechanism of gene exchange is transformation. It is
possible that the genes transferred via natural transformation, although at an
extremely low frequency below the levels of detection used in the
experiments reported here. Other bacteria closely related to B. cepacia have
been shown to transfer their genes via transformation. P. stutzeri and A.

calcoaceticus and Neisseria are all members of the beta Proteobacteria, (like B.

171

cepacia) and are naturally competent. Transformation has been observed in
B. cepacia (Mark Frischer, personal communication). Transformation is
important in nature in the evolution of new pathways (Stewart and Carlson,
1986). Although this event occurred slowly in vitro on a geological time
scale this could represent an important source of gene transfer in nature.
Infrequent events in small areas become frequent events when extrapolated

over large areas. For example, Frischer et a1. (1994) have calculated that a rate

of 10’15 transformants per liter per ng DNA per recipient per day reported for

marine bacteria becomes 1011 transformation events per year when

extrapolated to all of Tampa Bay Estuary in Florida.

‘ Transduction is another possible explanation for the transfer event.
Almost 50% of Burkholderia cepacia strains examined are susceptible to
lysogenic phage, and approximately 10% of B. cepacia strains carry transducing
phage (Matsumoto, 1986). Although two experiments were done to test for
transduction (data not shown) neither yielded positive results. This should
not be weighed too heavily since a positive control was not done.
Mechanism of integration: How the genes from RASC integrated into
their new host chromosome is unknown. Integration of 2,4-D genes into the
chromosome has been reported by Ka who observed loss of a 41 kb 2,4-D
plasmid in from Alcaligenes paradoxus strain 2118P with concomitant
insertion of a fragment that hybridized to the tfdA gene probe into the
chromosome of this organism that could still degrade 2,4-D (Ka, et al., 1994).
A plasmid nearly identical to this, pKA4, which was isolated from a strain
isolated from the same plot as 2118C at the Kellogg Biological Station, is
similar to pJP4, in that it is also an Inc P18 plasmid, and also hybridizes at

172

medium stringency to tfdA and tde from pJP4. Because of the similarity to
the pJP4 plasmid, this type of integration is probably different from that of
strain RASC, which contains no plasmids similar to pJP4.

Perhaps these genes integrated into a site in B. cepacia D5 which is
homologous to that of RASC. One potential candidate for homologous DNA
fragments are the repetitive extragenomic palindromic (REP) sequences. REP
sequences are highly conserved, inverted repeats of about 18 bp which appear
to be ubiquitous in gram negative bacteria. Although the presence of REP
sequences has been known since 1981, (Higgins, et al., 1982) the function of
these sequences is uncertain (Lupski and Weinstock, 1992; deBruijn, et al.,
1993; Higgins, et al., 1988). There has been speculation that REP sequences
may be involved in gene conversions and gene rearrangements (Higgins, et
al., 1988; Stern, et al., 1984; Gilson, 1987), or that they may have the potential
for transposition (Stern,1984) and are probably important for the evolution of
bacterial chromosomes (Lupski, 1992). Since the REP fragments from RASC
transferred to D5 along with the 2,4-D genes, it is tempting to speculate that
the REP sequences may have played some role in the integration of RASC
DNA into the D5 chromosome. There have been unpublished reports that
recombination occurred between the REP sequences on a plasmid and on the
chromosome (Stern, et al., 1984.) Other reports have indicated that
chromosomal duplications of the histidine operon in E. coli are located
between REP sequences. These REP fragments are not found within the
coding region of genes (Stern, 1984) but directly adjacent to them. E. coli
contains between 500-1000 copies of REP sequences which occupy up to one °/o
of the total genome (Higgins, 1988; Stern, 1984). Although REP sequences,

173

like Alu sequences of humans, may be shown to have other primary
functions, they still may be sites of recombination.

Another poSsibility mechanism for integration into the chromosome
may come from transposons, although the few experiments performed here
do not support this possibility. Since the transferred DNA was stably
integrated into the genome of strain RASC and the transgenic strain
D5(RASC) this indicates that the genes are not on an unstable transposon, as
is the case for the 3-chlorobenzoate transposon Tn 5271 which resides on
plasmid pBRC60 ( N akatsu, et al., 1991; Fulthorpe and Wyndham, 1991 ) or
the toluene transposon on plasmid pWWO Chakrabarty, et al., 1978; Tsuda
and Ieno, 1987; Burlage, et al., 1989).

There is evidence that the 2,4-D genes in this strain are not on located
on a highly mobile transposon, since self mobilizing plasmids introduced
into RASC failed to carry the 2,4-D genes from RASC. This, however, still
does not rule out the possibility that these genes are flanked by low frequency
mobilizing insertion sequences.

Additional observations: Strain RASC did not hybridize to the tde or tde
gene probes from pJP4 indicating that it has some type of gene which is less
than 55% similar in nucleotide homology. RASC and the transgenic strains
also failed to hybridize to the cchBC gene cluster of pAC27 which encodes
the degradative genes for 3—chlorobenzoate degradation. RASC probably
contains a dichlorocatechol dioxygenase, since the 2,4-D pathway in strain
RASC proceeds by the ortho pathway (Maltseva, personal communication).
This would be a unique dichlorocatechol dioxygenase gene since the three
known chlorocatechol dioxygenase genes (from tde, cch, and tch) all share

174

at least 53% amino acid similarity (van der Meer, et al., 1992). Perhaps the
ring cleavage enzyme in RASC is similar to 3,4-dihydroxybenzoate
(protocatechuate) dioxygenase enzyme, which has been described in B. cepacia
(Walsh, 1983). This enzyme has been shown to degrade 6-chloro
protocatechuate (Walsh, 1983) and may conceivably be the enzyme
responsible for further catabolism of 3,5-dichlorocatechol.

Since there were no probes available to determine whether genes from
the lower pathway of 2,4-D degradation transferred from RASC, it is not
known whether the entire pathway transferred or only the first two genes
transferred. Either case is possible. Since at least 15 kb of DNA from RASC
transferred, it is feasible for the lower pathway, including genes encoding ring
cleavage enzymes, to have been transferred as well. However, it is equally
likely that the remaining genes could have already been present in B. cepacia
since this organism also contains the B-keto adipate (ortho) cleavage pathway,
which is required for the lower pathway of 2,4-D degradation in pJP4. The
recipient strain, B. cepacia D5, is derived from an ancestral strain which
contains the 3,4-dihydroxybenzoate (protocatechuate) dioxygenase. enzyme
(Walsh and Ballou, 1983).

Perhaps one of the reasons for the catholic appetite of B. cepacia is the
many IS elements it contains, which could conceivably be rearranged in new
ways to accommodate new compounds (Lessie, 1990). It is widely accepted
that the B-ketoadipate pathway has evolved to accept many different aromatic
compounds (Omston, et al., 1990). B. cepacia, formerly more aptly named
Pseudomonas multivorans , has the reputation of degrading more types of

compounds than any other known species (Stanier, et al., 1966; Lessie, et al.,

175

1990). The genetic plasticity and nutritional versatility of this strain may

explain why it can accommodate genes for 2,4-D degradation.

In summary, the tfdA and tde genes from strain RASC were shown to
be transferred to and expressed in B. cepacia D5. Although gene exchange
occurred only very slowly, it was reproducible, and transgenic strains with
very stable 2,4-D phenotypes were obtained. This transfer of 2,4-D genes in
vitro supports the observation seen in the phylogenetic congruency test
(chapter 4) in which identical tfdA genes are found in phylogenetically

diverse strains.

ACKNOWLEDGEMENTS

I thank Dr. Yuichi Suwa for his probes from strain RASC and for useful
comments. Special thanks are given to Dr. Roberta Fulthorpe for helpful
discussions and to Dr. Larry Forney for encouragement. I also thank the
people in the Research on Microbial Evolution Laboratory for discussions and
ideas. This work was supported by the National Science Foundation grant
number BIR912006.

BIBLIOGRAPHY

BIBLIOGRAPHY

Allewell, N. (1989). ”Evolving to dissirnilate hydrocarbons.” TIBS 14: 473-474.

Altschul, S. F., W. Gish, W. Miller, E. W. Myers and D. J. Lipman (1990).
”Basic Local Alignment Search Tool.” J. Mol. Biol.» 215: 403-410.

Amann, R., W. Ludwig and K.-H. Schleifer (1994). ”Identification of
uncultured bacteria: a challenging task for molecular taxonomists.” ASM
News 60(7): 360-365.

Amy, P. S., I. W. Schulke, L. M. Frazier and R. J. Seidler (1985).
”Characterization of aquatic bacteria and cloning of genes specifying partial
degradation of 2,4-dichlorophenoxyacetic acid.” Appl. Env. Microbiol. 49(5):
1237-1245.

Ausmees, N. R. and A. L. Heinaru (1990). ”New plasmids of herbicide 2,4—D
degradation.” Genetica 26(4): 770-772.

Bagdasarian, M., R.Lurz, B. Ruckert, F. C. H. Franklin, M. M. Bagdasarian,
].Frey and K. N. Timmis (1981). ”Specific-purpose plasmid cloning vectors. II.
broad host range, high copy number, RSF 1010-derived vectors, and a host-
vector system for gene cloning in Pseudomonas.” Gene 16: 237-247.

Bakkeren, 6., Z. Koukolikova-Nicola, N. Grimsley and B. Hohn (1989).
”Recovery of Agrobacterium tumefaciens T-DNA molecules from whole
plants early after transfer.” Cell 57: 847-857.

Bale, M. 1., M. 1. Day and]. F. Fry (1988). ”Novel method for studying plasmid
transfer in undisturbed river epilithon.” Appl. Environ. Microbiol. 54(11):
2756-2758.

Bale, M. 1., I. C. Fry and M. I. Day (1987). ”Plasmid transfer between strains of
Pseudomonas aeruginosa on membrane filters attached to river stones.” J.
Gen. Microbiol. 133: 3099-3107.

Bale, M. 1., J. C. Fry and M. 1. Day (1988). ”Transfer and occurence of large

mercury resistance plasmids in river epilithon.” Applied and Env. Microbiol.
54(4): 972-978.

176

177

Barns, S. M. R. E. Fundyga, M. W. Jeffries, and N. R. Pace. (1994). "Remarkable
archael diversity detected in a Yellowstone National Park hot spring
environment." Proc. Natl. Acad. Sci. 91:1609-1613.

Baron, M., D. G. Norman and I. D. Campbell (1991). ”Protein modules.” TTBS
16: 13-17.

Baum, D. (1992). ”Phylogenetic species concepts.” TREE 7(1).

Begley, S. and T. Waldrop (1989). ”Microbes to the rescue.” Newsweek (June
19): 56-57.

Bell, G. R. (1957). ”Some morphological and biochemical characteristics of a
soil bacterium which decomposes 2,4-dichlorophenoxyacetic acid.” Can. J.
Microbiol. 3: 821-840.

Bennett, P. M. and]. Grinsted, Eds. (1988). Detection and use of transposons.
Methods in Microbiology. London, Academic Press.

Bentjen, S. A., J. K. Fredrickson, P.VanVoris and S. W. Li (1989). ”Intact soil-
core microcosms for evaluating the fate and ecological impact of the release of
genetically engineered microorganisms.” Appl. and Env. Microbiol. 55(1): 198-
202.

Berg, D. E. and M. M. Howe, Eds. (1989). Mobile DNA. Washington,D.C.,
American Society for Microbiology.

Bertrand, J. C., M. Almallah, M. Acquaviva and G. Mille (1990).
”Biodegradation of hydrocarbons by an extremely halophilic
archaebacterium.” Let. Appl. Microbiol. 11: 260-263.

Bhat, M. A., M. Tsuda, K. Horiike, M. N ozaki, C. S. Vaidyanathan and T.
Nakazawa (1994). ”Identification and characterization of a new plasmid
carrying genes for degradation of 2,4-dichlorophenoxyacetate from
Pseudomonas cepacia CSV90.” ABM 60(1): 307-312.

Bleakley, B. H. and D. L. Crawford (1988). ”The effects of varying moisture and
nutrient levels on the transfer of a conjugative plasmid between
Streptomyces species in soil.” Can. J. Microbiol. 35: 544-549.

Bogosian, G. and I. F. Kane (1991). Fate of recombinant Escherichia coli K-12
strains in the environment. in Advances in Applied Microbiology. Academic
Press. pp 87-131.

178

Bollag, 1. M., C. S. Helling and M. Alexander (1968). ”2,4-D metabolism.” 1.
Agr. Food Chem. 16(5): 826-833.

Bounds, H. C. and A. R. Colmer (1964). ”Detoxification of some herbicides by
streptomyces.” Weeds : 249-252.

Bradley, D. E. and P. A. Williams (1982). ”The TOL plasmid is naturally
derepressed for transfer.” 1. Gen. Microbiol. 128: 3019-3024.

Brisson-Noel, A., M. Arthur and P. Courvalin (1988). ”Evidence for natural
gene transfer from Gram-positive cocci to Escherichia coli.” Jour. Bact. 170:
1739-1745.

Brokamp, A. and F. R. J. Schmidt (1991). ”Survival of Alcaligenes
xylosoxidans degrading 2,2-dichloropropionate and horizontal transfer of its
halidohydrolase gene in a soil microocosm.” Curr. Microbiol. 22: 299-306.

Brosius, 1., M. L. Palmer, P. 1. Kennedy and H. F. Noller (1978). ”Complete
nucleotide sequence of a 16S ribosomal RNA gene from Escherichia coli.”
Proc. Natl. Acad. Sci. 75(10): 4801-4805.

Bruggen, A. H. C. v., K. N. 10chimsen, E. M. Steinberger, P. Segers and M.
Gillis (1993). ”Classification of Rhizomonas suberifaciens, an unnamed
Rhizomonas species, and Sphingomonas spp. in rRNA superfamily IV.”
International Journal of Systematic Bacteriology 43(1): 1-7.

Burlage, R. S., L. A. Bemis, A. C. Layton, G. S. Sayler and F. Larimer (1990).
”Comparative genetic organization of incompatibility group P degradative
plasmids.” 1. Bacteriol. 172(12): 6818-6825.

Burlage, R.S. , S. W. Hooper, and G. S. Sayler. (1989). "The TOL (pWWO)
catabolic plasmid." Appl. Environ. Microbiol. 55:1323-1328.

Busse, H.-1., T. el-Banna, H. Oyaizu and G. Auling (1992). ”Identification of
xenobiotic-degrading isolates from the beta subclass of the proteobacteria by a
polyphasic approach including 165 rRN A partial sequensing.” Internatioanl
Journal of Systematic Bacteriology 42(1): 19-26.

Calabrese, V. G. M., W. E. Holben, B. M. Schroeter, N. L. Tonso and A. 1.
Sextone (1991). Assessement of diversity of 2,4-D-catabolizing plasmids in
bacteria isolated from different soils. Agronomy Society of America Annual
Meeting, Denver, Colorado,

179

Cano, R. 1. (1994). ”Bacillus DNA in amber: a window to ancient symbiotic
relationships?” ASM News 60(3): 129-134.

Chakrabarty, A. M. and I. C. Gunsalas (1971). ”Degradative pathways specified
by extrachromosomal gene clusters in Pseudomonads.” Genetics 68(1): 510.

Chapman, P. 1. (1976). ”Microbail degradation of halogenated compounds.”
Biochem. Soc. Trans. 4: 463-466.

Chatterjee, D. K. and P. Chatterjee (1987). ”Expression of degradative genes of
Pseudomonas putida in Caulobacter crescentus.” Jour. Bacteriol. 169(7): 2962-
2966.

Chatterjee, D. K., S. T. Kellogg, D. R. Watkins and A. M. Chakrabarty (1988).
Plasmids in the biodegradation of chlorinated aromatic compounds. in
Plasmids Ed. Levy.

Chatterjee, D. K.,]. 1. Kilbane and A. M. Chakrabarty (1982). ”Biodegradation
of 2,4,5-trichlorophenoxyacetic acid in soil by a pure culture of Pseudomonas
cepacia.” Appl. Env. Micriobiol. 44(2): 514-516.

Chaudry, G. R. and G. H. Huang (1988). ”Isolation and characterization of a
new plasmid from a Flavobacterium sp. which carries the genes for
degradation of 2,4-dichlorophenoxyacetate.” 1. Bacteriol. 170(9): 3897-3902.

Chaudry, G. R., G. A. Toranzos and A. R. Bhatti (1989). ”Novel method for
monitoring genetically engineered microorganisms in the environment.”
Appl. Env. Microbiol. 55(5): 1301-1304.

Chen, S. and M. Alexander (1989). ”Reasons for the acclimation for 2,4-D
biodegradation in lake water.” 1. Environ. Qua]. 18: 153-156.

Cheng, H.-P. and T. G. Lessie (1994). ”Multiple replicons constituting the
genome of Pseudomonas cepacia 17616.” 1. Bact. 176(13): 4034-4042.

Christie, P. 1. and R. Z. Korman (1987). ”Two conjugation systems associated
with Streptococcus faecilis plasmid pCF10. Identification of a conjugative
transposon that transfers between S. faecilis and B. subtilis.” 1. Bact. 169: 2529-
2536.

Clewell, D., Ed. (1993). Bacterial Conjugation. New Yorks, Plenum Press.
Colwell, R. K. (1985). ”Viable but non-culturable Vibrio cholerae and related

pathogens in the environment: implication for the release of genetically
engineered microorganisms.” Bio/ technology 3: 817-820.

180

Corbett, J. R., K. Wright and A. C. Baille (1984). The Biochemical Mode of
Action of Pesticides. London, Academic Press.

Cork, D. J. and 1. P. Krueger (1991). Microbial transformation of herbicides and
pesticides. Advances in Applied Microbiology Eds. S. E. Neidleman and A. I.
Laskin. San Diego, Academic Press. 364.

Coughter,1. P. and G. J. Stewart (1980). ”Genetic exchange in the
environment.” Antonie van Leeuwenhoek 55: 15-22.

Creti, R., E. Ceccarelli, M. Bocchetta, A. M. Sanangelantoni, O. Tiboni and P.
Palm (1994). ”Evolution of translational elongation factor (EF) sequences:
reliability of global phylogenies inferred from EF-lalpha(Tu) and EF-2(G)
proteins.” Proc. Natl. Acad. Sci., USA 91: 3255-3259.

Czelusniak,1., M. Goodman, N. D. Moncrief and S. M. Kehoe (1990).
Maximum parsimony approach to construction of evolutionary trees from
aligned homologous sequences. Molecular Evolution: Computer analysis of
protein and nucleic acid sequences Ed. R. F. Doolittle. San Diego, Academic
Press. 736.

Dagley, S. (1976). ”Microbial catabolism and the carbon cycle.” Biochem. Soc.
Trans. 4: 455-458.

Datta, N. and R. W. Hedges (1972). ”Host ranges of R factors.” Jour. Gen.
Microbio. 70: 453-460.

Davidson, M. W., M. J. Page and M. F. French (1991). ”Drugs for Bugs.” Am.
Lab. 23(12): 34-338. ’

De Vos, R, M. Goor and J. D. Ley (1985). ”Ribosomal ribonucleic acid cistron
similarities of phytopathogenic Pseudomonas species.” International Journal
of Systematic Bacteriology 1985(2): 169-184.

deBruijn, F. J. (1992). ”Use of repetitive (repetitive extragenic palindromic and
enterobacterial repetitive intergeneric consensus)sequences and the
polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti
isolates and other soil bacteria.” Applied and Environmental Microbiology
58(7): 2180-2187.

181

Dewhirst, F. E., B. 1. Faster and P. L. Bright (1989). ”Chromobacterium,
Eikenella, Kingella, Neisseria, Simonsiella, and Virteoscilla species comprise
a major branch of the beta group of Proteobacteria by 16S ribosomal
ribonucleic acid sequence comparison: tranfer of Eikenella and Simsonsiella
to the family Neisseriaceae (emend.).” International Journal of Systematic

Bacteriology 39(3): 258-266.

Dewhirst, F. E., B. J. Paster, 1. Olsen and G. 1. Fraser (1992). ”Phylogeny of 54
representative strains of species in the family Pasteurellaceae as determinedby
comparison of 16S rRNA.” J. Bacteriol. 174(6): 2002-2013.

Dickerson, R. E. (1980). ”Evolution and gene transfer in purple photosynthetic
bacteria.” Nature 283: 210-212.

Dobzhansky, T. (1973). ”Nothing in biology makes sense except in the light of
evolution.” American Biology teacher 25: 125-129.

Don, R. H. and J. M. Pemberton (1981). ”Properties of six pesticide degradation
plasmids isolated from Alcaligenes paradoxus and Alcaligenes eutrophus.”
Jour. Bacteriol. 145(2): 681-686.

Don, R. H. and J. M. Pemberton (1985). ”Genetic and physical map of the 2,4-
dichlorophenoxyacetic acid-degrading plasmid pJP4.” J. Bacteriol. 161(1): 466-
468.

Don, R. H., A. J. Weightman and H. J. K. a. K. N. Timmis (1985). ”Transposon
mutagenesis and cloning of the pathways for degradation of 2,4-
dichlorophenoxyacetic acid and 3-chlorobenzoate in Alcaligenes eutrophus

JMPlB4(pJP4).” :

Doolittle, R. F., Ed. (1990). Molecular Evolution: Computer Analysis of
Protein and nucleic Acid Sequences. Methods in Enzymology. San Diego,
Academic Press.

Doolittle, R. F., D. F. Feng, K. L. Anderson and M. R. Alberro (1990). ”A
naturally occuring horizontal gene transfer from a eukaryote to a prokaryote.”
J. Mol. Evol. 31: 383-388.

Doolittle, W. F. (1988). ”Bacterial evolution.” Can. J. Microbiol. 34: 547-551.

Doolittle, W. F. and J. R. Brown (1994). ”Tempo, mode, the progenote, and the
universal root.” Proc. Natl. Acad. Sci. USA 91: 6721-6728.

Doolittle, W. F. and C. Sapienza (1980). ”Selfish genes, the phenotype
paradigm and genome evolution.” Nature 284: 601-603.

182

Drake, J. W. (1991). ”A constant rate of spontaneous mutation in DNA-based
microbes.” Proc. Natl. Acad. Sci. USA 88: 7160-7164.

DuBose, R. F., D. E. Dykhuizen and D. Hartl (1988). ”Genetic exchange among
natural isolates of bacteria: recombination within the phoA gene of
Escherichia coli.” Proc. Natl. Acad. Sci. 85: 7036-7040.

Duxbury,1. M., J.T'iedje, M. Alexander and J. E. Dawson (1970). ”2,4-D
metabolism: enzymatic conversion of chloromalylacetic acid to succinic acid.”
JAgr Food Chem 18(2): 199-201.

Dykhuizen, D. E. and L. Green (1991). ”Recombination in Escherichia coli and
the definition of biological species.” J. Bact. 173(2): 7257-7268.

Eberhard, W. G. (1990). ”Evolution in bacterial plasmids and levels of
selection.” Quarterly Rev. Biol. 61(1): 3-22.

Eckhardt, T. (1978). ”A rapic method for the identification of plasmid
desoxyribonucleic acid in bacteria.” Plasmid 1: 584-588.

Eden, P. A., T. M. Schmidt, R. P.1Blakemore and N. R. Pace (1991).
”Phylogenetic analysis of Aquaspirillum magnetotactium using polymerase
chain reaction-amplified 16S rRNA-specific DNA.” International Jouranl of
Systematic Bacteriology 41(2): 324-325.

Elsas, J. D. V., J. T. Trevors, M. E. Starodub and L. S. V. Overbeek (1990).
”Transfer of plasmid RP4 between pseudomonads after introduction into soil;
influence of spacial and temporal aspects of inoculation.” FEMS Microbiol.
Ecol. 73: 1-12.

Fallenleaf (1990). Proceedings of the Fallen Leaf LAke Conference on
Promiscuous Plasmids in Gram-Negative and -Positive Bacteria, Fallen Leaf
Lake, South Lake Tahoe, California, Plasmid.

Felsenstein,1. (1981). ”Evolutionary trees from DNA sequences: a maximum
likelihood approach." Journal of Molecular Evolution 17: 368-376.

Felsenstein,1. (1985). ”Confidence limits on phylogenies: an approach using
the bootstrap.” Evolution 39(4): 783-791.

Felsenstein,1. (1988). ”Phylogenies from molecular sequences: Inference and
reliability.” Annu. Rev. Genet. 22: 521-565.

183

Felsenstein, J. (1992). ”Phylogenies from restriction sites: a maxirnum-
likelihood approach.” Evolution 46(1): 159-173.

Fisher, P. R., J. Appleton and J. M. P{emberton (1978). ”Isolation and
characterization of the pesticide-degrading plasmid pJPl from Alcaligenes
paradoxus.” J. Bacteriol. 135(3): 798-804.

Flores, E. and C. P. Walk (1985). ”Identification of facultatively heterotrophic,
N2-fixing cyanobacteria able to receive plasmid vectors from Escherichia coli
by conjugation.” J. Bacteriol. 162(3): 1339-1341.

Fox, G. B., E. Stackebrandt, R. B. Hespell, 1. Gibson, 1. Maniloff, T. A. Dyer, R. S.
Wolfe, et al. (1980). ”The phylogeny of prokaryotes.” Science 209: 457-463.

Fox, G. B., J. D. Wisotzkey and P. Jurtshuk (1992). ”How close is close: 16S
rRNA sequence identity may not be sufficient to guarantee species identity.” :

Fox, 1. L. (1983). ”Soil microbes pose problems for pesticides.” Science 221:
1029-1031.

Fox, 1. L. (1989). ”Native microbes'role in Alaskan clean-up.” Bio / Technology
7: 852.

Frey, J. and M. Bagdasarian (1989). The molecular biology of Ian plasmids.
Promiscuous Plasmids of Gram-Negative Bacteria . Academic Press Limited.
79-94.

Friedrich, B., M. Meyer and H. Schlegel (1983). ”Transfer and expression of the
herbicide-degrading plasmid pJP4 in aerobic autotrophic bacteria.” Arch.
Microbiol. 134: 92-97.

Fries, M. R., J. Zhou, J. Chee-Sanford and J. M. Tiedje (1994). ”Isolation,
characterization, and distribution of denitrifying toluene degraders from a
variety of habitats.” Appl. Environ. Microbiol. 60(8): 2802-2810.

Frischer, M. E., G. J. Stewart and J. H. Paul (submitted). ”Plasmid transfer to
indigenous marine bacterial populations by natural transforamtion.”

Frisher, M. B., J. M. Thurmond and J. H. Paul (1990). ”Natural plasmid
transformation in a high-frequency-of-transformation marine Vibrio strain.”
Applied and Environmental Microbiology 56(11): 3439-3444.

184

Fry,1. C. and M. 1. Day (1990). Bacterial Genetics in Natural Environments.
Cambridge, Chapman and Hall.

Fukuda, M. (1993). ”Diverstiy of chloroaromatic oxygenases.” Current Biology
4: 339-343.

Fukumori, F. and R. P. Hausinger (1993). ”Alcaligenes eutrophus JMP134 "2,4-
Dichlorophenoxyacetate monooxygenase" is an alpha-ketoglutarate—
dependent dioxygenase.” J. Bacteriol. 175(7): 2083-2086.

Fulthorpe, R. and R. C. Wyndham (1992). ”Involvement of a chlorobenzoate-
catabolic transposon, Tn 5271, in community adaptation to chlorobiphenyl,
chloroaniline, and 2,4-dichlorophenoxyacetic acid in a freshwater ecosystem.”

Appl. Environ. Microbiol. 58(1): 314-325.

Fulthorpe, R. R., N. A. Straus and R. C. Wyndham (1989). Bacterial adaptation
to chlorobenzoate contamination in the Niagara region investigated by
DNAzDNA colony hybridization. In Aquatic Toxicology and Environmental
Fate. Eds. G. W. S. 11 and M. A. Lewis. Philadelphia, American Society for
Testing and Materials. 59-71.

Fulthorpe, R. R. and R. C. Wyndham (1991). ”Transfer and expression of the
catabolic plasmid pBR60 in wild bacterial recipients in a freshwater
ecosystem.” Appl. Environ. Microbiol. 57(5): 1546-1553.

Fulthorpe, R. R., C. McGowan, O. V. Maltseva, W. H. Holben, and J. M.
Tiedje. (1995) 2,4-Dichlrophenoxyacetic acid degrading bacteria are mosaics of
catabolic genes. (submitted)

Ghisalba, O. (1983). ”Chemical wastes and their biodegradation - an
overview.” Experientia : 1247-1257.

Ghosal, D. and I.-S. You (1988). ”Nucleotide homology and organization of
chlorocatechol oxidation genes of plasmids pJP4 and pAC27.” Mol. Gen.
Genet. 211: 113-120.

Ghosal, D., I.-S. You, D. K. Chatterjee and A. M. Chakrabarty (1985). ”Genes
specifying degradation of 3-chlorobenzoic acid in plasmids pAC27 and pJP4.”
Proc. Nat. Aca. Sci. USA 82: 1638-1642.

Ghosal, D., I.-S. You, D.K.Chatterjee and A. M. Chakrabarty (1985). ”Microbial
degradation of halogenated compounds.” Science 228(4696): 135-142.

185

Gibson, D. T., Ed. (1984). Microbial Degradation of Organic Compounds.
Microbiology Series. New York, Marcel Dekker.

Gilson, B., J.-M. Clement, D. Brutlag and M. Hofnung (1984). ”A family of
dispersed repetitive extragenic DNA sequences in E. coli.” EMBO 3(6): 1417-
1421.

Gilson, E., J. M. Clément, D. Perrin and M. Hofnung (1987). ”Palindromic
units: a case of highly repetitive DNA sequences in bacteria.” Trends in
Genetics 3(8): 226-230.

Giovannoni, S. 1., S. Turner, G. 1. Olsen, S. Barns, D. J. Lane and N. Pace (1988).
”Evolutionary relationships among cyanobacteria and green chloroplasts.” J.
Bacteriol. 170: 3584-3592.

Gish, W. and D. 1. States (1993). ”Identification of protein coding regions by
database similarity search.” Nature Genetics 3: 266-272.

Goodfellow, M., D. Jones and F. G. Priest, Eds. (1985). Computer-assisted
Bacterial Systematics. Special Publication of the Society for Gneral
Microbiology. London, Academic Press.

Graham, J. B. and C. A. [stock (1978). ”Genetic exchange in Bacillus subtilis in
soil.” Mol. Gen. Gen. 166: 287-290.

Graham, J. B. and C. A. [stock (1979). ”Gene exchange and natural selection
cause Bacillus subtilis to evolve in soil culture.” Science 204: 637-639.

Greer, C. W., J. Hawari and R. Samson (1990). ”Influence of envrionmental
factors on 2,4-dichlorophenoxyacetic acid degradation by Pseudomonas
cepacia isolated from peat.” Arch. Micriobiol 154: 317-322.

Greer, L. E., J. A. Robinson and D. R. Shelton (1992). ”Kinetic comparison of
seven strains of 2,4-dichlorophenoxyacetic acid-degrading bacteria.” Appl.
Environ. Microbiol. :

Guerry, P., D. J. LeBlanc and S. Faldow (1973). ”General method for the
isolation of plasmid deoxyribonucleic acid.” J. Bact. 116(2): 1064-1066.

Harayama, S. and R. H. Don (1985). Catabolic plasmids: their anaysis and
utilization in the manipulation of bacterial metabolic activities. Genetic
Engineering Principles and Methods Eds. J. K. Setlow and A. Hollander. New
York, Plenum Press. 283-307.

186

Harayama, S. and M. Kok (1992). ”Functional and evolutionary relationships
among diverse oxygenases.” Annu. Rev. Microbiol. 46: 565-601.

Hardy, K. G., Ed. (1987). Plasmids: A Pracitical Approach. The Practical
Approach Series. Oxford, IRL Press.

Hardy, K. G. (1987). Purification of plamids. Plasmids: A Practical Approach .
Oxford, IRL Press. 180. -

Harker, A. R. and Y. Kim (1990). ”Trichloroethylene degradation by two
independent aromatic degrading pathways in Alcaligenes eutrophus JMP134.”
Appl. Env. Microbiol. 56(4): 1179-1181.

Harker, A. R., R. H. Olsen and R. J. Seidler (1989). ”Phenoxyacetic acid
degradation by the 2,4-Dichlorophenoxyacetic acid (TFD) pathway of plasmid
pJP4: mapping and characterization of the T'FD regulatory gnen, tde.” J.
Bacteriol. 171(1): 314-320.

Harrigan, W. F. (1970). Probability tables for the estimation of microbial
numbers by the multiple tube technique. Laboratory Methods in Food and
Dairy Microbiology . London, Harcourt, Brace Jovanovich.

Hartnett, C., E. L. Neidle, K.-L. Ngai and L. N. Omston (1990). ”DNA
sequences of genes encoding Acinetobacter calcoaceticus protocatechuate 3,4-
dioxygenase: evidence indicating shuffling of genes and of DNA sequences
within genes during their evolutionary divergence.” Journal of Bacteriology
172(2): 956-966.

Hazen, T. C. and L. Jimenez (1988). ”Enumeration and identification of
bacteria from environmental samples using nucleic acid probes.” Microbiol.
Sciences 5(11): 340-343.

Heinemann,1. A. (1991). ”Genetics of gene transfer between species.” TTG 7(6):
181-185.

Heinemann, J. A. and G. F. Sprague (1989). ”Bacterial conjugative plasmids
mobilize DNA transfer between bacteria and yeast.” Nature 340: 205-209.

Helling, C. S., J.-M. Bollag and J. E. Dawson (1968). ”Cleavage of ether-oxygen
bond in phenoxyacetic acid by an Arthrobacter species.” J. Agric. Food Chem.
16(3): 538-539.

Hensche, R. B. and F. R. J. Schmidt, Eds. (1990). Use of wide hst range
promoters to monitor the fate of recombinant DNA in soil. ‘

187

Henschke, R. B. and F. R. J. Schmidt (1990). ”Plasmid mobilization from
genetically engineered bacteria to members of the indigenous soil microflora
in situ.” Curr. Microbiol. 20: 105-110.

Higgins, C. F., R. S. McLaren and S. F. N ewbury (1988). ”Repetitive
extragenomic palindromic sequences, mRN A stability and gene expression:
evolution by gene conversion? - a review.” Gene 72: 3-14.

Hiraishi, A. (1992). ”Direct automated sequencing of 16S rDNA amplified by
polymerase chain reaction from bacterial cultures without DNA
purification.” Letters in Applied Microbiology 15: 210-213.

Hiraishi, A., Y. Hoshino and T. Satoh (1991). ”Rhodoferax fermentans gen.
nov., sp. nov., a phototrophic purple nonsulfur bacterium previously
referred to as the "Rhodocyclus gelatinonus-like" group.” Arch. Microbiol.
155: 330-336.

Holben, W. E., V. G. M. Calabrese, D. Harris, J. O. Ka and J. M. Tiedje (1993).
Analysis of structure and selection in microbial communities by molecular
methods. Trends in Microbial Ecology Eds. R. Guerrero and C. Pedros-Alio.
Spanish Society for Microbiology. 367-370.

Holben, W. B., V. G. M. Calabrese, D. Harris, J. O. Ka and J. M. Tiedje (1993).
”Analysis of structure and selection in microbial communities by molecular
methods.” Trends in Microbial Ecology : 367-370.

Holben, W. E. ,.J K. Jansson, B. K. Chelm andJ. M. Tiedje (1988). ”DNA probe
method for the detection of specific microorganisms in the soil bacterial
community.” Appl. Environ. Microbiol. 54: 703-711.

Holben, W. E., B. M. Schroeter, V. G. M. Calabrese, R. H. Olsen, 1. K. Kukor, V.
O. Biederbeck, A. E. Smith, et al. (1992). ”Gene probe analysis of soil microbial
population selected by amendment with 2,4-dichlorophenoxyacetic acid.”
Appl. Env. Microbiol. 58(12): 3941-3948.

Istock, C. A., K. E. Duncan, N. Ferguson and X. Zhou (1992). ”Sexuality in a
natural population of bacteria - Bacillus subtilus challenges the clonal
paradigm.” Molecular Ecology 1: 95-103.

Jacob, F. (1977). ”Evolution and tinkering.” Science 196: 1161-1166.
Jacobsen, C. S. and J. C. Pedersen (1992). ”Growth and survival of

Pseudomonas cepacia DB01(pROlOl) in soil amended with 2,4-
dichlrophenoxyacetic acid.” Biodegradation 2: 245-252.

188

Jacobson, C. S. and J. C. Pedersen (1992). ”Mineralization of 2,4-
dichlorophenoxyacetic acid (2,4-D) in soil inoculated with Pseudomonas
cepacia DBOl(pR0101) and Alcaligenes eutrophus JMP134(pJP4): effects of
inoculation level and substrate concentration.” Biodegradation 2: 253-263.

Jain, R. K., G. S. Sayler, J. T. Wilson, L. Houston and D. Pacia (1987).
”Maintenance and stability of introduced genotypes in groundwater aquifer
material.” Appl. Env. Microbiol. 53(5): 996-1002.

Jansson, J. (1988). Development of gene probe methodology for detection of
specific bacteri in environmental samples. PhD Dissertastion. Michigan State
University.

Jeenes, D. J. and P. A. Williams (1982). ”Excision and integration of
degradative pathway genes from TOL plasmid pWWO.” J. Bact. 150(1): 188-
194.

Johnson, J. L. and N. 1. Palleroni (1989). ”Deoxyribonucleic acid similarities
among Pseudomonas species.” International Journal of Systematic
Bacteriology 39(3): 230-235.

Ka, J. 0., W. E. Holben and J. M. Tiedje (1994a). ”Analysis of competition in
soil among 2,4-dichlorophenoxyacetic acid-degrading bacteria.” Appl.
Environ. Microbiol. 60(4):

Ka, J. 0., W. E. Holben and J. M. Tiedje (1994b). ”Genetic and phenotypic
diversity of 2,4-dichlorophenoxyacetic acid (2,4-D)-degrading bacteria isolated
from 2,4-D-treated field soils.” Appl. Environ. Microbiol. 60(4):

Ka, J. 0., W. E. Holben and J. M. Tiedje (1994c). ”Use of gene probes to aid in
recovery and identification of functionally dominant 2,4-
Dichlorophenoxyacetic acid-degrading populations in soil.” Appl. Environ.
Microbiol. 60(4):

Ka,1. O. and J. M. Tiedje (1994d). ”Integration and excision of a 2,4-
dichlorophenoxyacetic acid-degradative plasmid in Alcaligenes paradoxus
and evidence of its natural intergeneric transfer.” 1. Bact. 176(17): 000-000.

Kado, C. I. and S.-T. Liu (1981). ”Rapid procedure for detection and isolation of
large and small plasmids.” J. Bacteriol. 145(3): 1365-1373.

Kamagata, Y. (1994). personal communication.

189

Kaphammer, B., J. 1. Kukor and R. H. Atlas (1990). ”Regulation of tdeDEF by
tde of the 2,4-dichlorophenoxyactic acid degradation plasmid pJP4.” J.
Bacteriol. 172(5): 2280-2286.

Kellogg, S. T., D. K. Chatterjee and A. M. Chakrabarty (1981). ”Plasmid-assisted
molecular breeding: New technique for enhanced biodegradation of persistent
toxic chemicals.” Science 214: 1133-1135.

Kilbane, J. J., D. K. Chatterjee and A. M. Chakrabarty (1983). ”Detoxification of
2,4,5-trichlorophenoxyacetic acid from contaminated soil by Pseudomonas
cepacia.” Appl. Env. Microbiol. 45(5): 1697-1700.

Kimura, M. (1983). The neutral theory of molecular evolution. In: Evolution
of genes and proteins Eds. M. Nei and R. K. Koehn. Sunderland, MA, Sinauer
Associates.

Klingmiiller, W. (1991). ”Plasmid transfer in natural soil: a case by case study
' with nitrogen-fixing Enterobacter.” FEMS Microbial Ecology 85: 106-116.

Kobayashi, H. and B. E. Rittmann (1982). ”Microbial removal of hazardous
organic compounds.” Environ. Sci. Techno. 16(3): 170-181.

Kozyreva, L. P., Y. V. Shurukhin, Z. I. Finkel'shtein, B. P. Baskunov and L. A.
Golovleva (1993). ”Metabolism of the herbicide 2,4-D by a Nocardioides
simplex strain.” Microbiology 62(1): 78-84.

Krishnapillai, V. (1988). ”Molecular genetic analysis of bacterial plasmid
promiscuity.” FEMS Microbiol. Rev. 54: 223-238.

Kukor, J. 1., R. H. Olsen and J.-S. Siak (1989). ”Recruitment of a
chromosomally encoded malylacetate reductase fro degradation of 2,4-
dichlorophenoxyatecit acid by plasmid pJP .” J. Bacteriol. 171(6): 3385-3390.

Kumar, S., K. Tamura and M. Nei (1993). MEGA: Molecular Evolutionary
Genetics Analysis, version 1.01. University Park, PA, The Pennsylvania State
University.

Kuramitzu, H. K. and M. J. Casadaban (1986). ”Transposition of the gram
positive transposon TN917 in E. coli.” J.Bact. 167: 711-712.

Laguerre, G., M.-R. Allard, F. Revoy and N. Amarger (1994). ”Rapid
identification of Rhizobia by restriction fragment length polymorphis analysis
of PCR-amplified 16$ rRNA genes.” ABM 60(1): 56-63.

190

Lake, 1. A. (1991). ”Tracing origins with molecular sequences: metazoan and
eukaryotic beginnings.” TTBS 16: 46-50.

Lane, D. 1., B. Pace, G. 1. Olsen, D. A. Stahl, M. Sogin and N. R. Pace (1985).
”Rapid determination of 165 ribosomal RNA sequences for phylogenetic
analyses.” P.N.A.S. 82: 6955-6959.

Lang, B. F. (1984). ”The mitochondrial genome of the fission yeast 5. pombe:
highly homologous introns are inserted in the same positions of the
otherwise less conserved Cox 1 genes in S. pombe and Apergillus nidulans.”
EMBO 3: 2129-2136.

Latorre,1., W. Reineke and H.-1. Knackmuss (1984). ”Microbial metabolism of
chloroanilines: enhanced evolution by natural genetic exchange.” Arch.
Microbiol. 140: 159-165.

Lenski, R. E. and M. Travisano (1994). ”Dynamics of adaptation and
diversification: a 10,000-generation experiment with bacterial populations.”
Proc. Natl. Acad. Sci. USA 91: 6808-6814.

Lerner, S. A., T. T. Wu and E. C. C. Lin (1964). ”Evolution of a catabolic
pathway in bacteria.” Science 146: 1313-1315.

Lessie, T. G., M. S. Wood, A. Byrne and A. Ferrante (1990). Transposable gene-
activating elements in Pseudomonas cepacia. In: Pseudomonas:
Biotransformations, Pathogenesis, and Evolving Biotechnology Ed. Silver.
Washington, DC, American Society for Microbiology.

Levy, S. B. and B. M. Marshall (1988). Genetic transfer in the natural
environment. In: The Release of Genetically-Engineered Microorganisms.
New York, Academic Press. 61-76.

Levy, S. B. and R. V. Miller (1989). Gene transfer in the environment. New
York, McGraw Hill Publishing Comapany.

Lewin, R. (1989(?)). ”Fish to bacterium gene transfer.” Science 227: 1020.

Ley, J. d., P. Segers, K. Kersters, W. Mannheim and A. Lievens (1986). ”Intra-
and intergeneric similarities of the Bordatella ribosomal ribonucleic acid
cistrons: proposal for a new family, Alcaligenaceae.” Interantional Journal of
Systematic Bacteriology 36(3): 405-414.

Li, W.-H. (1986). ”Evolutionary change of restriction cleavage sites and
phylogenetic inference.” Genetics 113: 187-213.

191

Li, X., M. Dorsch, T. D. Dot, L. I. Sly, E. Stackebrandt and A. C. Haywood (1993).
”Phylogenetic studies of the rRNA group II pseudomonads based on 165
rRNA gene sequences.” Journal of Applied Bacteriology 74: 324-329.

Loos, M. A., R. N. Roberts and M. Alexander (1967). ”Formation of 2,4-
dichlorophenol and 2,4-dichloroanisole from 2,4-dichlorophenoxyacetate by
Arthrobacter sp.” Can. J. Microbiol. 13: 691-699.

Loos, M. A., R. N. Roberts and M. Alexander (1967). ”Phenols as intermediates
in the decomposition of phenoxyacetates by an Arthrobacter species.” Can. J.
Microbiol. 13: 679-690.

Lorenz, M. G., B. W. Aardema and W. Wackernagel (1988). ”Highly efficient
genetic transformation of Bacillus subtilis attached to sand grains.” 1. Gen.
Microbiol. 134: 107-112.

Lorenz, M. G. and W. Wackemagel (1991). ”High frequency of natural genetic
transformation of Pseudomonas stutzeri in soil extract supplemented with a
carbon/ energy and phosphorus source.” Appl. Env. Microbiol. 57(4): 1246-
1251.

Lupski,1. R. and G. M. Weinstock (1992). ”Short, interspersed repetitive DNA
sequences in prokaryotic genomes.” J. Bacteriology 174(14): 4525-4529.

Macrina, F. L., D. J. Kopecko, K. R. Jones, D. J. Ayers and S. M. McCowen
(1978). ”A multiple plasmid containing Escherichia coli strain: Convenient
source of size reference plasmid mlecules.” Plasmid 1: 417-420.

Maddison, W. P. and D. R. Maddison (1992). MacClade: Analysis of phylogeny
and character evolution. Version 3.0. Sunderland, MA, Sinauer Associates.

Madsen, E. L. (1991). ”Determining in situ biodegradation: facts and
challenges.” Environ. Sci. Technol. 25(10): 1663-1672.

Mae, A. A., R. O. Maritis, N. R. Ausmees, V. M. Koiv and A. L. Heinaru
(1993). ”Characterization of a new 2,4-dicl'tlorophenoxyacetic acid degrading
plasmid pEST4011: physical map and localization of catabolic genes.” Journal
of General Microbiology 139: 3165-3170.

Mancini, P., S. Fertels, D. Nave and M. Gealt (1987). ”Mobilization of plasmid
pHSV106 from Escherichia coli HB101 in a laboratory-scale waste treatment
facility.” Appl. Environ. Microbiol. 53(4): 665-671.

192

Maniatis, T., E. F. Fritsch, J. Sambrook (1982). Molecular Cloning: a Laboratory
Manual. Cold Spring Harbor Laboratory Press. New York.

Margulis, L. (1991). Symbiosis in Evolution: Origins of cell motility. In:
Evolution of Life: Fossils, Molcules and Culture ,Eds. S. Osawa and T. Honjo.
Tokyo, Springer-Verlag. 305-324.

Margulis, L. (1992). ”Biodiversity: molecular biological domains, symbiosis
and kingdom origins.” BioSystems 27: 39-51.

Marshall, E. (1991). ”A is for apple, Alar and alarmist?” Science 254: 20-22.

Martin, J. P. and I. Fridovitch (1981). ”Evidence for a natural gene transfer

from the ponyfish to its bioluminescent symbiont Photobacter leignathi.”
Jour. Biol. Chem. 256(12): 6080-6089.

Massol-Deya, A. A., D. A. Odelson, R. F. Hickey and J. M. Tiedje (in press).
Bacterial community fingerprinting of amplified 16S and 16-238 ribosomal
DNA gene sequences and restriction endonuclease analysis (ARDRA). In:
Molecular Microbial Ecology Manual, Ed. F. deBruijn. Kluwgr pub.

Matsumoto, H., Y. Itoh, S. Ohta and Y. Terawaki (1986). ”A generalized
transducing phage of Pseudomonas cepacia.” 1. Gen. Microbiol. 132: 2583-2586.

Mazodier, P., R. Petter and C. Thompson (1989). ”Intergenic conjugation
between Escherichia coli and Streptomyces species.” Jour. Bacteriology 171(6):
3583-3585.

McCall, P. J., S. A. Vrona and S. S. Kelley (1981). ”Fate of uniformly carbon-14
ring labeled 2,4,5-trichlorophenoxyacetic acid and 2,4,-dichlorophenoxyacetic
acid.” J. Agric. Food. Chem. 29(1):

McClure, N. C., J. C. Fry and A.J.Weightman (1991). ”Survival and catabolic
activity of natural and genetically engineered bacteria in a labolratory-scale
activated sludge unit.” Appl. Env. Microbiol. 57(2): 366-373.

McClure, N. C., A. J. Weightman and J. C. Fry (1989). ”Survival of
Pseudomonas putida UWCl containing cloned catabolic genes in a model
activated-sludge unit.” Appl. Environ. Microbiol. 55(10): 2627-2634.

Mergeay, M., D. Springael and E. Top (1990). Gene transfer in polluted soils.
In: Bacterial Genetics in Natural Environments , Eds. J. C. Fry and M. 1. Day.
London, Chapman and Hall. 259.

193

Meulien, P. and P. Broda (1982). ”Identification of chromosomally integrated
TOL DNA in cured derivatives of Pseudomonas putida PAW1.” J. Bacteriol.
152(2): 911-914.

Michel, F. and B.Dujon (1986). ”Genetic exchanges between bacteriophage T4
and filamentous fungi?” Cell 46: 3231.

Milkman, R. and M. M. Bridges (1990). ”Molecular evolution of the
Escherichia coli chromosome. III Clonal fragments.” Genetics 126: 505-517.

Milkman, R. and A. Stoltzfus (1988). ”Molecular evolution of the Escherichia
coli chromosome. H Clonal segments.” Genetics 120: 359-366.

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor,
Cold Spring Harbor Laboratory.

Miller, R. V., T. A. Kokjohn and G. S. Sayler (1990). Environmental and
molecular characterization of systems which affect genenome alteration in
Pseudomonas aeruginosa. In: Pseudomonas: Biotransformations,
Pathodenesis, and Evolving Biotechnology Ed. Silver. Washington, DC,
American Society for Microbiology. 252-268.

Moore, E. R. B., R. Wittich, P. Fortnagel and K. N. Timmis (1993). ”16S
ribosomal RNA gene sequence characterization and phylogenetic analysis of a
dibenzo-p-dioxin-degrading isolate within the new genus Sphingomonas.”
Letters in Applied Microbiology 17: 115-118.

Moran, N. A., M. A. Manson, P. Bauman and H. Ishikawa (1993). ”A
molecular clock in endosymbiotic bacteria calibrated using the insect hosts.”
Proc. R. Soc. Lond. B. 253: 167-171.

Morgan, P. and R. 1. Watkinson (1989). ”Microbiological methods for the
cleanup of soil and ground water contaminated with halogenated organic
compounds.” FEMS Microbiol. Rev. 63: 277-300.

Mortlock, R. P., Ed. (1992). The Evolution of Metabolic Function. Boca Raton,
CRC Press.

Mortlock, R. P. and M. A. Gallo (1992). "Experiments in the evolution of
catabolic pathways usig modern bacteria." In: The Evolution of Metabolic
Function Ed. R. P. Mortlock. Boca Raton, CRC Press. 1-13.

194

Mayer, C. L., F. C. Dobbs and D. M. Karl (1994). ”Estimation of diversity and
community structure through restriction fragment length polymorphism
distribution analysis of bacterial 16S rRNA genes from a microbial mat at an
active, hydrothermal vent system, Loihi Seamount, Hawaii.” Applied and
Environmental Microbiology 60(3): 871-879.

Murray, R. G. E., D. J. Brenner, R. R. Colwell, P. D. Vos, M. Goodfellow, P. A.
D. Grimont, N. Pfennig, et al. (1990). ”Report of the Ad Hoc Committee on
approaches to taxonomy within the Proteobacteria.” International Journal of
Systematic Bacteriology 40(2): 213-215.

Nakatsu, C., J. Ng, R. Singh, N. Staius and C. Wyndham (1991).
”Chlorobenzoate catabolic transposon Tn5271 is a composite class I element
with flanking class II insertion sequences.” Proc. Natl. Acad. Sci. In Press,:

Nakatsu, C. H. (1992). The biochemical pathway and catabolic genes for
chlorobenzoate degradation encoded on the transopson Tn 5271 of
Alcaligenes sp. strain BR60. PhD Thesis, Carleton University.

Natarajan, M. R. and P. Oriel (1991). ”Conjugal transfer of recombinant
transposon Tn916 from Escherichia coli to Bacillus stearothermophilus.”
Plasmid 26: 67-73.

Nei, M. and W.-H. Li (1979). ”Mathematical model for studying genetic
variation in terms of restriction endonucleases.” Proc. Natl. Acad. Sci. 76(10):
5269-5273.

Neilson, J. W., K. L. Josephson, S. D. Pillai and I. L. Pepper (1992). ”Polymerase
chain reaction and gene probe detection of the 2,4-dichlorophenoxyacetic acid
degradation plasmid, pJP4.” Applied and Environmental Microbiology 58(4):
1271-1275.

NiiBlein, K., D. F. Dwyer, D. Maris and K. Timmis (In press). ”Expression and
transfer of catabolic pathways harbored by Pseudomonas spp. introduced into
activated sludge microcosms.” :

Oh, K.-H. and O. H. Tuovinen (1990). ”Degradation of 2,4-
dichlorophenoxyacetic acid by mixed cultures of bacteria.” J. Indust. Microbiol.
6: 275-278.

Olsen, G. 1., R. Overbeek, N. Larsen, T. L. Marsh, M. J. McCaughey, M. A.
Maciukenas, W.-M. Kuan, et al. (1992). ”The Ribosomal Database Project.”
Nucleic Acids Research 20: 2199-2200.

195

Olsen, G. J., C. R. Woese and R. Overbeek (1994). ”The winds of
(Evolutionary) change: breathing new life into microbiology.” J. Bact. 176(1): 1-
6.

Orgel, L. E. and F. H. C. Crick (1980). ”Selfish DNA: the ultimate parasite.”
Nature 284: 604-607.

Omston, L. N., J. Houghton, E. L. Neidle and L. A. Gregg (19). Subtle selection
and novel mutation during evolutionary divergence of the B—ketoadipate
pathway.

Omston, L. N. and D. Parke (1976). ”Evolution of catabolic Pathways.”
Biochem. Soc. Trans. 4: 468-473.

Omston, L. N. and W. K. Yeh (1979). ”Origins of metabolic diversity:
evolutionary divergence by sequence repitition.” Proc. Natl. Acad. Sci. 76(8):
3996-4000.

Pallernoni, N. 1., R. W. Ballard, E. Ralston and M. Doudoroff (1972).
”Deoxyribonucleic acid homologies among some Pseudomonas species.” J.
Bacteriol. 110(1): 1-11.

Palleroni, N. 1., R. Kunisawa, R. Contopoulou and M. Doudoroff (1973).
”Nucleic Acid Homologies in the genus Pseudomonas.” Int. J. Sys. Bacteriol.
24(4): 333-339.

Faster, B. 1., F. E. Dewhirst, 1. Olsen and G. 1. Fraser (1994). ”Phylogeny of
Bacteroides, Prevotella, and Porphyromonas spp. and related bacteria.” 1. Bact.
176(3): 725-732.

Pemberton, J. M. (1981). ”Genetic engineering and biological detoxification of
environmental pollutants.” Residue Reviews 78: 1-11.

Pemberton, J. M., B. Corney and R. H. Don (1979). Evolution and spread of
pesticide degrading ability among soil micro-organisms. Plasmids of Medical,
Environmental and Commercial Importance, Spitzingsee, F.D.G., Elsevier.

Perkins, E. 1., M. P. Gordon, 0. Caceres and P. F. Lurquin (1990). ”Organization
and sequence analysis of the 2,4-dichlorophenol hydroxylase and
dichlorocatechol oxidative operonsa of plasmid pJP4.” J.Bacteriol. 172(5): 2351-
2359.

196

Pertsova, R. N., F.Kunc and L. A. Golovleva (1984). ”Degradation of 3-
chlorobenzoate in soil by pseudomonads carrying biodegradative plasmids.”
Folia Microbiol. 29: 242-247.

Pfennig, N. (1984). Microbial behavior in natural environments. In: The
Microbes Eds. D. P. Kelly and N. G. Carr. Cambridge, Cambridge University
Press.

Pinkney, M. and C. M. Thomas (1987). ” Replication and maintenace of
promiscuous plasmids of Gram-negative bacteria.” Microbial. Sciences 4(6):
186-191.

Piruzyan, L. A., A. G. Malenkav and G. M. Barrenboym (1985)." Chemistry
and the biosphere." In: Global Ecology Ed. C. H. Southwick. Sunderland,
Massachusetts, Sinauer Associates. 323.

Pastel, S. (1988). "Controlling toxic chemicals." In: IEd. L. R. Brown. New
York, W. W. Norton. 237.

Reineke, W., D. Jeenes, P. A. Williams and H.-J. Knackmuss (1982). ”TOL
plasmid pWWO in constructed halabenzaate-degrading Pseudomonas
strains: prevention of meta pathway.” J. Bact. 150(1): 195-210.

Reineke, W. and H.-J. Knackmuss (1988). ”Microbial degradation of
haloaromatics.” Ann. Rev. Microbial. 42: 263-287.

Reineke, W. and H. J. Knackmuss (1979). ”Construction of haloaromatics
utilising bacteria.” Nature 277(1): 385-386.

Reineke, W. and H. J. Knackmuss (1980). ”Hybrid pathway for chlorobenzoate
metabolism in Pseudomonas sp. 813 derivatives.” 1. Bacteriol. 142: 467-473.

Roberts, L. (1987). ”Discovering microbes with a taste for PCBs.” Science 237:
975-977.

Roberts, M. S. and F. M. Cohan (1993). ”The effect of DNA sequence
divergence on sexual isolation in Bacillus.” Genetics 134: 401-408.

Roberts, R. C., R. Burioni and D. K. Helsinki (1990). ”Genetic characterization
of the stabilizing hmctions of a region of broad-host-range plasmid RK2.” 1.
Bacterial. 172(11): 6204-6216.

Rochkind, M. L., J. W. Blackburn and G. Sayler, Eds. (1986). Microbial
Decomposition of Chlorinated Aromatic Compounds. Cincinnati, US. E.P.A.

197

Rasenshine, 1., R. Tchelet and M. Mevarech (1987). ”The mechanism of DNA
transfer in the mating system of an Archaebacterium.” Science (245): 1387-
1389.

Raszak, D. B. and R. R. Calwell (1987). ”Survival strategies of bacteria in the
natural environment.” Microbiol. Rev. 51: 365-379.

Saunders, 1. R., A. Dacherty and G. O. Humphreys (1984). ”Transfarrnation of
bacteria by plasmid DNA.” Meth. Microbiol 17: 61-95.

Saye, D. 1., O. Ogunseitan, G. S. Sayler and R. V. Miller (1987). ”Potential for
transduction of plasmids in a natural freshwater environment: effect of
plasmid donor concentration and a natural microbial community on
transduction in Speudomonas aeruginosa.” Appl. Environ. Microbiol. 53(5):
987-995.

Saye, D. 1., 0. A. Ogunseitan, G. S. Sayler and R. V. Miller (1990).
”Transduction of linked chromosomal genes between Pseudomonas
aeruginaosa strains during incubation in situ in a freshwater habitat.” Appl.
Environ. Microbiol. 56(1): 140-145.

Sayler, G. S., S. W. Hooper, A. C. Layton and J. M. H. King (1990). ”Catabolic
plasmids of environmental and ecological significance.” Micrab. Ecol 19: 1-20.

Schilf, W. and W. Klingmiiller (1983). ”Experiments with Escherichia coli on
the dispersal of plasmids in environmental samples.” Recombinant DNA
Technical Bulletin 6(3): 101-102.

Schliimann, M. (1991). Enzyme und gene des Abbaus mono- und
dichlorsubstituierter Brenzkatechine: Woher kommt ein neuer Abbauweg?
Bialogischer Abbau von Chlarkohlenwasserstaffen Ed. B. Weigert. Berlin,
Technishe Universitéit Berlin. 87-107.

Schliimann, M., D. H. Pieper and H. J. Knackmuss (1990). Enzymes of
haloaromatic degradation: variation of Alcaligenes on a theme by
Pseudomonas. In: Psuedomonas: Biotransformations, pathogenesis, and
evolving biotechnology. Eds, Silver, S., A. M. Chakrabarty, B. Iglewski and S.
Kaplan, Washington, DC, American Society of Microbialogists.

Schliimann, M., E. Schmidt and H. J. Knackmuss (1984). ”Different types of a
key enzyme of halaaromatic degradation in benzaate utilizing bacteria.” Appl.
Sys. Microbiol. 5: 259.

198

Schmidt, T. M., E. F. DeLang and N. R. Pace (1991). ”Analysis of a marine
picoplankton community by 16S rRNA gene cloning and sequencing.” 1.
Bacterial. 173(14): 4371-4378.

Schofield, P. R., A. H. Gibson, W. F. Dudman and J. M. Watson (1987).
”Evidence for genetic exchange and recombination of Rhizobium symbiotic
plasmids in a soil population.” Appl. Env. Microbial. 53(12): 2942-2947.
Short, K. A., J. D. Doyle and R. J. King (1991). ”Effects of 2,4-Dichlorophenol, a
metabolite of a genetically engineeered bacterium, and 2,4-
dichloraphenoxyacetate on some microorganism-mediated ecological
processed in soil.” Appl. Env. Microbial. 57(2): 412-418.

Sikorski, R. S., W. Michaud, H. L. Levin, J. D. Baeke and P. Heiter (1990).
”Trans-kingdom promiscuity.” Nature 345: 581-582.

Silver, S., A. M. Chakrabarty, B. Iglewski and S. Kaplan, Eds. (1990).
Psuedomonas: Biotransformations, pathogenesis, and evolving
biotechnology. 'Washington, DC, American Society of Microbialogists.

Simon, R., U. Priefer and A. Puehler (1983). ”A broad host range mobilization
system for in viva genetic engineering: transposon mutagenesis in gram
negative bacteria.” Bio/ Tech November: 784-791.

Simon, R., U. Priefer and A. Piihler (1983). Vector plasmids for in-viva and
in-vitro manipulation of gram-negative bacteria. Molecular Genetics of the
Bacteria-Plant Interaction Ed. A. Piihler. Berlin, Springer-Verlag.

Sinclair, M. 1., P. C. Maxwell, B. R. Lyon and B. W. Holloway (1986).
”Chromosomal location of TOL plasmid DNA in Pseudomonas putida.” J.
Bact. 168:

Siuda, J. F. and J. F. DeBernardis (1973). ”Naturally occurring halogenated
organic compounds.” :

Slater, J. H. (1985). ”Gene transfer in microbial communities.” Microbial Ecol.
: 89-98.

Smith, C. A. and C. M. Thomas (1989). Relationships and evolution of IncP
Plasmids. In: Promiscuous plasmids of Gram-Negative Bacteria. Academic
Press Limited.

Smith, G. R. (1991). ”Conjugal recombination in E. coli: myth and
mechanisms.” Cell 64: 19-27.

1 99
Smith, J. M. (1989). ”Generating novelty by symbiosis.” Nature 341: 248-285.

Smith, J. M., C. G. Dawson and B. G. Spratt (1991). ”Localized sex in bacteria.”
Nature 349: 29-31.

Smith, J. M., N. H. Smith, M. O'Rourke and B. G. Spratt (1993). ”How clonal
are bacteria?” Proc. Natl. Acad. Sci., USA 90: 4384-4388.

Sogin, M. L. (1991). The phylogenetic significance of sequence diversity and
length variations in eukaryotic small subunit ribosomal RNA coding regions.
In: New Perspectives on Evolution. Wiley-Liss. 175-188.

Sanea, S. (1989). ”A new look at bacteria.” ASM News 55(11): 584-585.

Souza, V., T. T. Nguyen, R. R. hudson, D. Pifiero and R. E. Lenski (1992).
”Hierachical analysis of linkage disequilibrium in Rhizobium populations:
Evidence for sex?” Proc. Natl. Acad. Sci., USA 89: 8389-8393.

Spratt, B. G., L. D. Bowler, Quian-Yun Zhang, J. Zhou and 1. M. Smith (1992).
”Role of interspecies transfer of chromosomal genes in the evolution of

penicillin resistance in pathogenic and cammensal Neisseria species.” J. Mol.
Evol. 34: 115-125.

Stachel, S. E. and P. C. Zambryski (1989). ”Generic trans-kingdom sex?”
Nature 340: 190-191.

Stackebrandt, B., R. G. E. Murray and H. G. Triiper (1988). ”Proteobacteria
classis nov., a name for the phylogenetic taxon that includes the "purple
bacteria and their relatives".” International Journal of Systematic Bacteriology
38(3): 321-325.

Stanier, R. Y., N. J. Palleroni and M. Douderaff (1966). ”The aerobic
Pseudomonads: a taxonomic study.” J. Gen. Microbial. 43: 159-279.

Slatkin, M. (1985). ”Gene flaw in natural populations.” Ann. Rev. Ecol. Syst.
16:393-430.

Slatkin, M. and W. P. Maddison (1989). ”A cladistic measure of gene flow
inferred from the phylogenies of alleles.” Genetics 123: 603-613.

Steffan, R. 1., A. Breen, R. M. Atlas and G. S. Sayler (1989). ”Monitoring
geneically engineered microorganisms in freshwater microcosms.” J. Indust.
Microbiol. 4: 441-446.

200

Stern, M. 1., G. F.-L. Ames, N. H. Smith, E. C. Robinson and C. F. Higgins
(1984). ”Pepetitive extragenic palindromic sequences: a major component of
the bacteria genome.” Cell 37: 1015-1026.

Stewart, C.-B. (1993). ”The powers and pitfalls of parsimony.” Nature 361: 603-
607.

Stewart, G. J. (1986). ”The biology of natural transformation.” Ann. Rev.
Microbiol. 40: 211-235.

Stewart, G. J. and C. D. Sinigalliana (1990). ”Detection of horizontal gene
transfer by natural transformation in native and introduced species of bacteria
in marine and synthetic sediments.” Appl. Environ. Microbiol. 56(6): 1818-
1824.

Statzky, G. (1989). "Gene transfer among bacteria in soil." In: Gene transfer in
the Environment Eds. S. B. Levy and R. V. Miller. New York, McGraw-Hill
Publishing Company. 434.

Stotzky, G. and H. Babich (1986). ”Survival of, and genetic transfer by,
genetically engineered bacteria in natural environments.” Adv. Appl.
Microbiol. 31: 93-138.

Streber, W. R., K. N. Timmis and M. H. Zenk (1987). ”Analysis, cloning, and
high-level expression of 2,4-dichlorophenaxyacetate monaoxygenase gene
tfdA of Alcaligenes eutrophus JMP134.” 1. Bacterial. 169(7): 2950-2955.

Suwa, Y., A. D. Wright, F. Fukumori, K. A. Nummy, R. P. Hausinger, L. 1.
Barney, and W. E. Holben (1995). Gene cloning and sequence analysis of a
chromosomally encoded 2,4-D dichlaraphenoxyacetate monaoxygenase gene
tfdA of Alcaligenes eutrophus JMP134. 1. Bacterial. (submitted).

Swofford, D. L. (1991). PAUP: Phylogenetic Analysis Using Parsimony,
version 3.05. Champaign, Illinois, Illinois Natural History Survey.

Thomas, C. M., Ed. (1989). Promiscuous Plasmids of Gram-Negative Bacteria.
San Diego, Academic Press Limited.

Thomas, C. M. and C. A. Smith (1987). ”Incompatibility group P plasmids:
genetics, evolution, and use in genetic manipulation.” Ann. Rev. Microbiol.
41: 77-101.

201

Thamashaw, M. F., R.Nutter, K. Pastle, M. D. Chilton, F. R. Blattner,
A.Powell, M. P. Gordon, et a1. (1980). ”Recombination between higher plant
DNA and the Ti plasmid of Agrobacterium tumefaciens.” Proc. Natl. Acad.
Sci. USA 77(11): 6448-6452.

Tiedje, J. M., J. M. Duxbury, M. Alexander and J. E. Dawson (1969). ”2,4-D
metabolism: Pathway of degradation of chlorocatechols by Arthrobacter sp.” 1.
Agr. Food Chem. 17(5): 1021-1025.

Timmis, K. N. and A. Piihler (1979). Plasmids of Medical, Environmental and
Commercial Importance. In: Plasmids of Medical Environmental and
Commercial Importance, Spitzingsee, Elsevier.

Tonso, N., N. Iwashita and M. Fukuda (1994). Phylogenetic analysis and
aromatic degradative capability of several Rhadococci isolated from
contaminated soils. American Society of Microbiology Meeting, Las Vegas.
Poster Q 171.

Tonso, N. L., V. G. M. Calabrese and W. E. Holben (submitted). ”Polyphasic
characterization of a suite of bacterial isolates capable of degrading 2,4-D.”

Top, E., M. Mergeay, D. Springael and W. Verstraete (1990). ” Gene escape
model: transfer of heavy metal resistance genes from Escherichia coli to

Alcaligenes eutrophus on agar plates and in soil samples.” Appl. Environ.
Microbiol. 56: 2471-2479.

Top, E., I. d. Smet, W. Verstraete, R. Dijkmans and M. Mergeay (1994).
”Exogenous isolation of mobilizing plasmids from polluted sails and
sludges.” Appl. Environ. Microbiol. 60(3). 831-839.

Tarsvik, V., J. Goksoyr and F. L. Daae (1990). ”High diversity in DNA of soil
bacteria.” Appl. Environ. Microbiol. 56(3): 782-787.

Torsvik, V., K. Salte, R. Sarheim and J. Gaksayr (1990). ”Comparison of
phenotypic diversity and DNA heterogeneity in a population of soil bacteria.”
Appl. Environ. Microbial. 56(3): 776-781.

Tourova, T. P. and A. S. Antonov (1987). ”Identification of microorganisms by
rapid DNA-DNA hybridization.” Methods in Microbiology 19: 333-355.

Trevors,1. T. (1987). ”Survival of Escherichia coli donor, recipient, and
transconjugant cells in soil.” Water Air Soil Pollut. 34: 409-414.

202

Trevors, J. T., T. Barkay and W. Baurquin (1986). ”Gene transfer among
bacteria in osil and aquatic environments: a review.” Can. J. Microbiol. 33:
191-198.

Trevors, J. T. and G. Berg (1989). ”Conjugal RP4 transfer between
pseudomonads in soil and recovery of RP4 plasmid DNA from soil.” Syst.
Appl. Microbiol. 11: 223-227.

Trevors,1. T. and K. M. Oddie (1986). ”R-plasmid transfer in soil and water.”
Can. J. Microbiol 32: 610-613.

Trevors, J. T. and M. E. Staradub (1987). ”R-plasmid transfer in non-sterile
agricultural soil.” Syst. Appl. Microbiol. 9: 312-315.

Trieu-Cuot, P., C. Carlier, P. Martin and P. Caurvalin (1987). ”Plasmid transfer
by conjugation from Escherichia coli to Gram-positive bacteria.” FEMS
Microbiological Letters 48: 289-294.

US.EPA Pesticide Fact Handbook. Park Ridge, N .J., Noyes Data Corporation.

Valdés, A. M. and D. Pifiero (1992). ”Phylogenetic estimation of plasmid
exchange in bacteria.” Evolution 46(3): 641-656.

Van Bruggen, A. H. C., K. N. Jochimsen, E. M. Steinberger, A. Segers and M.
Gillis (1993). ”Classification of Rhizomonas suberifaciens, and unnamed
Rhizomonas species, and Sphingomonas spp. in rRNA Superfamily IV.”
Internatioanl Journal of Systematic Bacteriology 43(1): 1-7.

Van der Meer, J. R., W. M. de Vos, S. Harayama and A. J. B. Zehnder (1992).
”Molecular mechanisms of genetic adaptation to xenobiotic compounds.”
Microbiol. Rev. 56(4): 677-694.

Van der Meer,1. R., R. I. L. Eggen, A. 1. B. Zehnder and W. M. d. V08 (1991).
”Sequence analysis of the Pseudomonas sp. strain P51 tcb gene cluster, which

encode metabolism of chlorinated catechals: evidence for specialization of
catechol 1,2-dioxygenases for chlorinated substrates.” 1. Bact. 173(8): 2425-2434.

Van Elsas,1. D. and J. T. Trevors (1990). "Plasmid transfer to indigenous
bacteria in soil and rhizosphere: problems and perspectives." In: Bacterial
Genetics in Natrual Environments Eds. J. C. Fry and M. 1. Day. Cambridge,
Chapman and Hall. 188-199.

Verstraete, W. and E. Top (1992). Holistic Environmental Biotechnology. .

203

Vollmer, M. D., K. Stadler-Fritzsche and M. Schliimann (1993). ”Conversion
of 2-chloromaleylacetate in Alcaligenes eutrophus JMP134.” Arch. Microbiol.
159: 1-7.

Walsh, T. A. and D. P. Ballou (1983). ”Halogeneated protocatechuates as
substrates for protocatechuate dioxygenase from Pseudomonas cepacia.” The J.
Biol. Chem. 258(23): 14413-14421.

Ward, D. M., M. M. Bateson, R. Weller and A. L. Ruff-Roberts (preprint).
”Ribosomal RNA analysis of microorganisms as they occur in nature.”
Advances in Microbial Ecology :

Wayne, L. G., D. J. Brenner, R. R. Colwell, P. A. D. Grimont, O. Kandler, M. I.
Krichevsky, L. H. Moore, et al. (1987). ”Report of the Ad Hoc Committee on
reconciliation of approaches to bacterial systematics.” International Journal of
Systematic Bacteriology: 37(4): 463-464.

Weinberg, S. R. and G. Stotzky (1972). ”Conjugation and genetic
recombination of Escherichia coli in soil.” Soil Biol. Biochem. 4: 191-180.

Weisburg, W. G., S. M. Barns, D. A. Pelletier and D. J. Lane (1991). ”16S
ribosomal DNA amplification for phylogenetic study.” J. Bact. 173(2): 697-703.

Wellington, E. M. H., N. Cresswell and V. A. Saunders (1990). ”Growth and
survival of streptomycete inoculants and extent of plasmid transfer in sterile
and nonsterile sail.” Appl. Environ. Microbiol. 46(4): 1413-1416.

Wheatcroft, R., D. G. McRae and R. W. Miller (1990). ”Changes in the
Rhizobium meliloti genome and the ability to detect supercoiled plasmids
during bacteroid development." Molecular Plant-Microbe Interactions 3(1): 9-
17.

Wheelis, M. L., 0. Kandler and C. R. Woese (1992). ”On the nature of global
classification.” Proc. Natl. Acad. Sci. 89: 2930-2934.

Wickham, G. S. and R. M. Atlas (1988). ”Plasmid frequency fluctuation in
bacterial populations from chemically stressed soil communities.” Appl.
Environ. Microbial. 54(9): 2192-2196.

Willems, A., 1. d. Ley, M. Gillis and K. Kersters (1991). ”Comamodaceae, a new
family encompassing the acidovarans rRNA complex, including Variovorax
paradoxus gen. nov., comb. nov., for Alcaligenes paradoxus (Davis 1969).”
International Journal of Systematic Bacteriology 41(3): 445-450.

204

Willets, N. (1985). Plasmids. Genetics of Bacteria Eds. J. Scaife, D. Leach and A.
Galizzi. London, Academic Press. 165-195.

Willets, N. and C. Crowther (1981). ”Mobilization of the non-conjugative
Ian plasmid RSF1010.” Genet. Res. Camb 37: 311-316.

Williams, P. A. and M. J. Warsey (1976). ”Plasmids and catabolism.” Biochem.
Soc. Trans. 4: 466-468.

Wilson, A. C., S. S. Carlson and T. J. White (1977). ”Biochemical Evolution.”
Ann. Rev. Biochem. 46: 573-639.

Woese, C. (1987). ”Bacterial evolution.” Microbiol. Rev. 51: 221-271.

Woese, C. R. (1994). ”There must be a prakaryote somewhere: microbialogy's
search for itself.” Microbiol. Rev. 58(1): 1-9.

Woese, C. R., P. Blanz and C. M. Hahn (1984). ”What isn't a Pseudomonad:
the importance of nomenclature in bacterial classification.” System. Appl.
Microbial. 5: 179-195.

Woese, C. R., 1. Gibson and G. E. Fox (1980). ”Do genealogical patterns in
purple photosynthetic bacteria reflect interspecific gene transfer?” Nature 283:
212-216.

Wiihrman, K. and J. Tomiuk, Eds. (1993). Transgenic Organisms: Risk
Assessment of Deliberate Release. Advances in Life Sciences. Basel,
Birkhauser Verlag.

Wolfe, C. J. and M. G. Haywood (1991). ”Restriction fragment length
polymorphism analysis reveals high levels of genetic divergence among the
light organ symbionts of flashlight fish." Biol. Bull. 181: 135-143.

Wang, F. Y. K., E. Stackebrandt, J. K. Ladha, D. E. Fleischman, R. A. Date and 1.
A. Fuerst (1994). ”Phylogenetic analysis of Bradyrhizobium japonicum and
photosynthetic stem-nodulating bacteria from Aeschynomene species grown
in separated geographical regions.” Appl. Environ. Microbiol. 60(3): 940-946.

Wyndham, C., R. K. Singh and N. A. Straus (1988). ”Catabolic instability,
plasmid gene deletion and recombination in Alcaligenes sp. BR60.” Arch.
Microbiol. 150: 237-243.

Wyndham, R. C., A. E. Cashore, C. H. Nakatsu and M. C. Peel (1994).
”Catabolic transposons.” Biodegradation in press.

205

Wyndham, R. C. and N. A. Straus (1988). ”Chlorobenzoate catabolism and
interactions between Alcaligenes and Pseudomonas species from Bloody Run
Creek.” Arch. Microbiol. 150: 230-236.

Yabuuchi, E., Y. Kosaka, H. Oyaizu, I. Yano, H. Hotta, Y. Hashimota, T. Ezaki,
et al. (1992). ”Proposal of Burkholderia gen. nov. and transfer of seven species
of the genus Pseudomonas homology group II to the new genus, with the
type species Burkholderia cepacia (Palleroni and Holmes 1981) comb. nov.”

Microbiol. Immunol. 36(12): 1251-1275.

Yadav, J. S., R. E. Wallace and C. A. Reddy (submitted). ”Degradation of
chlorobenzenes by the white-rot fungus Phanerochaete chrysosporium.”

Yano, K. (1980). "Degradative plasmids: aspects of microbial evolution." In:
Molecular Breeding and Genetics of Applied Microorganisms Eds. K.
Sakaguchi and M. Okanishi. Tokyo, Kodansha Ltd.

Zilberman, D., A. Schmitz, G. Casterline, E. Lichtenberg and]. B. Siebert
(1991). ”The economics of pesticide use and regulation.” Science 253: 518-522.

Zuckerkandl, E. and L. Pauling (1965). ”Molecules as documents of
evolutionary history.” J. Theor. Biol. 8: 357-366.

Zylstra, G. J. and D. T. Gibson (1991). "Aromatic hydrocarbon degradation: a
molecular approach." In: Genetic Engineering: Principles and Methods Ed. 1.
K. Setlow. New York, Plenum Press. 183-203.

Zylstra, G. 1., R. H. Olsen and D. P. Ballou (1989). ”Cloning, expression, and
regulation of the Pseudomonas cepacia protocatechuate 3,4-dioxygenase
genes.” J. Bact. 171(11): 5907-5914.

Zylstra, G. J., R. H. Olsen and D. P. Ballou (1989). ”Genetic organization and
sequence of the Pseudomonas cepacia genes for the alpha and beta subunits of
protocatechuate 3,4-dioxygenase.” J. Bact. 171(11): 5915-5921.

(1990) ”Validation of the publication of new names and new combinations
previously effectively published outside the IJSB.” International Journal of
Systematic Bacteriology 40(3): 320-321.

(1992). ”Validation of the publication of new names and new combinations
previously effectively published outside the IJSB.” International Journal of
Systematic Bacteriology 42(1): 191-192.

nxcurcaN STATE UNIV. LIBRQRIES
lllllW1WllllllllllllllllllillW11111111111"
31293014107258